The Oral Tradition of the Early Buddhists

Piya Tan The Oral Tradition of the Early Buddhists

§ Topic § Topic The Pali Canon and Commentaries 5.2.2 Āgama as textual authenticity 5.2.3 The 4 great references for the Dharma 1 The oral tradition 5.2.4 The 6 modes of conveyances (otaraṇa hara) 1.1 Teaching and listening; the 45-year ministry 5.2.5 The 4 great references for the Vinaya 1.1.1 The 45-year ministry 5.2.6 A list of lists: Fig 5.2.6 (tree diagram) 1.1.2 The highest training 5.2.7 Meditation and the path pericope 1.2 Listening and seeing 5.2.8 An early list of teachings (Gethin) 1.2.1 Discipleship through listening (savana) 5.2.9 Numerical lists 1.2.2 Discipleship through seeing (dassana) (1) Kumāra,pañha (Khp 4.2) 1.2.3 Discipleship as Vinaya (2) Kumāra,pañha & the Mahā,pañha 1+2 2 Living the Dharma 5.3 Mātikā and mātikā,dhara (Gethin) 2.1 Narrative and lesson 5.3.1 The mātikā in the Abhidhamma 2.2 The Pali texts are very old 5.3.2 The mātikā outside the Abhidhamma 2.3 Experience and meditative stillness 5.3.3 The rise of the Abhidhamma 3 Reading and listening 5.4 Aṅga: Limbs of the Teacher’s teaching 3.1 Sutta study promotes mindfulness 5.4.1 The limbs of the teaching 3.2 Importance of the Nikāyas 5.4.2 Nav’aṅga satthu,sāsana: the 9 limbs 3.2.1 The early Buddhist texts (EBTs) (13) The 4 analytic skills (paṭisambhidā) 3.2.2 The 4 Nikāyas 5.4.3 From scripture to Scripture 3.3 The suttas in programming language 5.4.4 The Chinese Buddhist canon 3.3.1 Listening tools 5.4.5 The Āgamas in Chinese 3.3.2 Examples of programme mechanism 6 Nikāya: The sutta collections 3.3.3 Fractals 6.1 Āgama and nikāya 4 Oral tradition as literature 6.1.1 North and south 4.1 The earliest Buddhist texts 6.1.2 Spread of the early canon 4.1.1 The meaning of “text” 6.2 Ti,piṭaka 4.1.2 Evidence for oral transmission 6.2.1 Oral transmission 4.2 The sound nature of the early oral tradition 6.2.2 The reciters 4.2.1 Narrative or didactic? 6.2.3 Monastics: forest or urban? 4.2.2 The approach formula 6.2.4 The 5 nikāyas 4.2.3 Adaptability of the early oral tradition (1) The Dīgha.nikāya 4.2.4 Sound similarities: The formal sutta opening (2) The Majjhima.nikāya Fig 4.2.4 Sound features in sutta opening (3) The Saṁyutta.nikāya 4.2.5 Waxing syllable principle: Fig 4.2.5 (4) The Aṅguttara.nikāya 4.2.6 Repetition (Mark Allon) (5) The Khuddaka.nikāya 4.2.7 Verbatim repetition (Bh Analayo) 7 Living Buddha Dharma fully 4.2.8 Pericope variations 7.1 The middle way of truth and beauty 5 Genre, āgama, mātikā, aṅga 7.2 The complete sutta 5.1 Genres 7.3 Discipleship is the path 5.1.1 Evolution of the early oral tradition 7.4 Early Buddhism as common sense 5.1.2 The 7 genres in early Buddhism 7.4.1 Sameness of awakening: Sambuddha Sutta 5.2 Āgama: All the teaching in practice 7.4.2 The “Pali drift” 5.2.1 Āgama as authenticity and tradition 7.4.3 The spirit of early Buddhism

http://dharmafarer.org 1 SD 58.1 The Oral Tradition of the Early Buddhists

The Pali Canon and Commentaries This list of the Pali Texts and Commentaries is given in the Encyclopaedia of Buddhism (Ency Bsm 2:343).

Text Commentary Commentary Author

Visuddhi,magga (Vism) Vism Mahā,ṭīkā (VismMhṭ) Buddhaghosa

VINAYA PIṬAKA Vinaya Piṭaka (V) Samanta,pāsādikā (VA) Buddhaghosa?1 Pāṭimokkha (Pṭmk) Kaṅkhā,vitaraṇī (PṭmkA) Buddhaghosa?

SUTTA PIṬAKA Dīgha,nikāya (D) Sumaṅgala,vilāsinī (DA = Sv) Buddhaghosa Majjhima,nikāya (M) Papañca,sūdanī (MA = Ps) Buddhaghosa Saṁyutta,nikāya (S) Sār’attha-p,pakāsinī (SA = Sp) Buddhaghosa Aṅg’uttara,nikāya (A) Manoratha,pūraṇī (AA = Mp) Buddhaghosa Khuddaka,nikāya (1) Khuddaka,pāṭha (Khp) Param’attha,jotikā I (KhpA) Attr Buddhaghosa* (2) Dhammapada (Dh) Dhammapad’aṭṭhakathā (DhA) Attr Buddhaghosa* (3) Udāna (U) Param’attha,dīpanī I (UA) Dhammapāla (4) Iti,vuttaka (It) Param’attha,dīpanī II (ItA) Dhammapāla (5) Sutta,nipāta (Sn) Param’attha,jotikā II (SnA) Attr Buddhaghosa* (6) Vimāna,vatthu (Vv) Param’attha,dīpanī III (VvA) Dhammapāla (7) Peta,vatthu (Pv) Param’attha,dīpanī IV (PvA) Dhammapāla (8) Thera,gāthā (Tha) Param’attha,dīpanī V (ThaA) Dhammapāla (9) Therī,gāthā (Thī) Param’attha,dīpanī VI (ThA) Dhammapāla (10) Jātaka (J) Jātak’aṭṭhakathā (J or JA) Attr Buddhaghosa (11) Niddesa (Nm Nc) Saddhamma-p.pajjotikā (NmA NcA) Upasena (12) Paṭisambhidā,magga (Pm) Saddhamma-p,pakāsinī (PmA) Mahānāma (13) Apadāna (Ap: ThaAp, ThīAp) Visuddha,jana,vilāsinī (ApA) Unknown2** (14) Buddha,vaṁsa (B) Madur’attha,vilāsinī (BA) Buddhadatta (15) Cariyā,piṭaka (C) Param’attha,dīpanī VII (CA) Dhammapāla

ABHIDHAMMA PIṬAKA Dhamma,saṅgaṇī (Dhs) Attha,sālinī (DhsA = Asl) Buddhaghosa? Vibhaṅga (Vbh) Sammoha,vinodanī (VbhA) Buddhaghosa? Kathā,vatthu (Kvu) KvuA Puggala,paññatti (Pug) PugA Dhātu,kathā (Dhk) DhkA Pañca-p,pakaraṇ’aṭṭhakathā Buddhaghosa? Yamaka (Yam) YamA Paṭṭhāna (Pa) PaṭA

A single asterisk (*) denotes: (a) Culla,buddhaghosa as the author of DhA: Malalasekera, Pali Literature of Ceylon, 96 f, and of JA, KhA and SnA: Barua, Ceylon Lectures, 88 f and Law Buddhaghosa, 1946:60; for SnA: Adikaram, Early History of Buddhism in Ceylon, 1946:7 f. (b) Buddhaghosa II as the author of VA and Kaṅkhā,vitaraṇī. For the double asterisks (**), see colophon in Comys and Gandha,vaṁsa (JPTS 1886:59, 68).

1 The question mark “?”: Buddhaghosa was accepted as author but doubts have been expressed in recent times. 2 Gandha,vaṁsa ascribes ApA to Buddhaghosa.

2 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

1 The oral tradition

1.1 TEACHING AND LISTENING

1.1.1 The 45-year ministry

1.1.1.1 Traditionally—according to the Theravāda monastics and teachers, represented by the ethnic Buddhisms of Sri Lanka, Myanmar, Thailand, Cambodia and Laos—the Buddha was born in 623 BCE and died in 543 BCE.3 Contemporary Western scholars generally accept the Buddha’s final passing away (parinirvana;4 P parinibbāna) as being between 410 and 400 BCE, probably 404 BCE, which then gives his dates as 484-404 BCE.5 In the Mahā,parinibbāna Sutta (D 16), the Buddha tells us that he went forth at 29.6 Six years later, at 35, he awakened as the Buddha, which would be 449 BCE. He taught for 45 years, passing away at 80, in 404 BCE. None of his teachings were written down in his life-time, but they were well taught (svākkha- ta) to a great number of people all over the central Gangetic plain, the most developed part of ancient India.

1.1.1.2 During the early formative years of Buddhism, the Buddha carefully chose and taught mostly those who were ready to understand his experience of awakening. These early disciples listened to the Buddha, at first as individuals or in small groups. The Buddha’s first converts were often serious practitioners or were spiritually ready to hear the teaching. The earliest of the Buddha’s monastic followers were the group of 5 monks, and the youth Yasa and his 54 friends. They all became arhats just like the Buddha (the Buddha being the “1st arhat” in our time).7 Within the 1st year of his ministry, the Buddha was able to muster a sangha of 60 arhats. He at once sent them out on his “great commission” as Dharma missioners to spread the Buddha-word in 60 different directions, making his teaching the world’s 1st missionary faith.8 In other words, unlike Brahminism, where the brahmins or priests claimed their status to be “the way and the truth” to God, salvation etc, the Buddha’s message is that the path of awakening is within us, attainable through self-effort—and only through self-effort. By examining what we really are, we better understand our mind so that we can tame it and free it: this is called self-awakening.

1.1.2 The highest training

1.1.2.1 While Brahminism and other God-religions preach all kind of beliefs, prayers and rituals, and putting the teacher or God above oneself, the Buddha teaches that all views are mind-made and provi- sional. The best ritual, the highest prayer, is our own spiritual practice, of knowing our mind, taming it

3 Sinhala traditional dates are 624-544 BCE respectively. See SD 49.8b (12.1.3); cf Mahā,parinibbāna S (D 16,4.39 n), SD 9. 4 “Parinirvana” is an anglicized term, Skt parinirvāṇa. 5 According to Gombrich: “the Buddha died 136 years before Aśoka’s inauguration, which means 404 BC” (“Dat- ing the historical Buddha,” 1992:246). He estimates the margin of error to be 7 years before to 5 years after this date, ie, 411-399 (1992:244). He notes the uncertainty about Aśoka’s dates, widening the margin of error’s upper limit as 422 BCE. K R Norman: “If we take an average, then the date is c411 ± 11 BCE” (1999:467 = Coll Papers 7, 2001). See Oxford DB: date of the Buddha; L S Cousins, “The early development of Buddhist literature and language in India,” 2013:89-93. See Oxford Dictionary of Buddhism 2003: date of the Buddha. 6 D 16,5.27.4/2:151 (SD 9). 7 See Sambuddha S (S 33.58), SD 49.10; Pavāraṇā S (S 8.7), SD 49.11. 8 See The great commission (SD 11.2). http://dharmafarer.org 3 SD 58.1 The Oral Tradition of the Early Buddhists and freeing it. The Buddha declares himself to be only a “shower of the way” (akkhatā) (Dh 276).9 We can and must work out our own salvation, just as we began life by learning to walk on our own two feet, and now we are thinking about religion and reality in our own minds.

1.1.2.2 Since the purpose of the Buddhist life is basically that of knowing “suffering and its ending,”10 as practitioners, our task is to hear and live the teaching to reach the path in this life itself, that is, to attain streamwinning11 at least. We progress towards the path of awakening by, first, refining our body and speech, freeing them of grosser defilements that would chain us to samsara (the cycle of lives and deaths), especially keeping ourself from falling into the subhuman states of the animal, the preta and the hell- being.12 Refining our body and speech—the training in moral virtue (sīla,sikkhā)—forms the basis for the training in mental cultivation (samādhi,sikkhā), the honing of the mind as the tool for directly seeing into true reality that is impermanence, unsatisfactoriness and non-self. The cultivation of this vision embodies the training in wisdom (paññā,sikkhā), ripening in the rightness (sammatta) of right knowledge and right freedom, making us arhats.13

1.2 LISTENING AND SEEING

1.2.1 Discipleship through listening (savana)

1.2.1.1 Our progress in the 3 trainings—those in moral virtue, mental cultivation and wisdom [1.1.2.2] —is rooted in our willingness and diligence in listening and hearing. Listening is an external but active aspect of discipleship (taking up the training with the path as our life’s goal). Hearing is an inward, “deep” attention, spiritual transformation. There are at least 2 Pali verbs relating to this initial phase of Dharma-spirited discipleship, that is, sussūsaka, “making ourself right and ready to listen,” that is, by “lending our ear” (ohita,sota). This act of listening is called savana; hence, it is said that “the timely listening to the Dharma is a supreme blessing” (kālena dhamma-s,savanaṁ, etam maṅgalam uttamaṁ).14 Technically (as noted above), this act of listening (savana) is merely an external one, an initial or nascent act: we can, and often, forget what we have heard. Hence, there is a need for repeated listening, for which we would have expected a word such as anussavana (anu-, “after, following, repeated” + savana, “listening”), but it has the negative sense of “hearsay,” or at best, the neutral, sense of “report.” “Right listening” is not repeated listening but active listening or “deep listening.” It is an “inward listening,” one that hears the silence and sees the sense of the unspoken or inexpressible (which, understandably, usually takes time). It transforms us spiritually into a better person, even a noble individual of the path. Hence, a vital quality of discipleship is that of “listening to the Dharma attentively” (ohita,soto dhammaṁ suṇāti). [1.2.1.3] Such a person who “has well heard” the Dharma is said to be suta, “one who has heard, learned.” A disciple who “has heard much” is bahu-s,suta, that is, also remembers the teachings. Of the monk disci-

9 See SD 49.21 (2.1.4.4). 10 Anurādha S (S 22.86), SD 21.13; Alagaddûpama S (M 22,38), SD 3.13; SD 40a.1 (11.1.1). 11 See (Anicca) Cakkhu S (S 25.1), SD 16.7. 12 On these subhuman states, see SD 48.1 (5.2.4). 13 See SD 10.16 (10). 14 Maṅgala S (Khp 5,8/3; Sn 265/47).

4 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists ples who have heard much, the Buddha declares Ānanda as the foremost,15 and amongst the laywoman disciples, it is Khujj’uttarā.16 Amongst the lay disciples, the houselord Citta is wisest amongst Dharma teachers (dhamma,kathikānaṁ).17

1.2.1.2 One of the most beautiful Pali stock passages describes how the mind of faith is fully engaged with the Dharma, thus: “He listens to the Dharma, fully attentive, fully receptive, all ears, minding it, directing his whole mind [heart] to it” (aṭṭhi,katvā [aṭṭhiṁ katvā] manasi katvā sabba,cetaso samannā- haritvā ohita,soto dhammaṁ suṇāti). Literally, this reads: “he listens to the Dharma, driven by purpose [intent on the meaning], gives ear, fully attentive, directing his mind totally to it.”18 This phrase is analysed as follows: faculty (indriya) aṭṭhi,katvā [aṭṭhiṁ katvā] making (the Dharma, the goal) his purpose faith manasi katvā minding it effort sabba,cetaso samannāharitvā attending to it with all his mind [heart] mindfulness ohita,soto (“with ear directed,” from odahati19) all ears concentration dhammaṁ suṇāti he listens to the Dharma wisdom

Through this “deep” listening of the Dharma, especially when it is taught by the arhats, the audience will surely benefit in terms of cultivating their spiritual faculties (indriya) of faith (saddhā), effort (viriya), mindfulness (sati), concentration (samādhi) and wisdom (paññā). These will, in turn, help them progress to attain the path in this life itself. The faith of the audience inspires them with a deep sense of purpose. Effort arises in them even as they keep their mind focused on the teaching. With mindfulness, the audience receive the flow of Dhar- ma into their hearts since they are “all ears,” listening without any distraction. As a natural fruit of this mental openness and spiritual readiness, wisdom arises to them even as they listen, and grows in due course. A more detailed description, in practical terms, of how this deep listening of the Dharma helps in our spiritual development is by way of the stages of discipleship (sāvakatta), which we shall examine next.

1.2.1.3 The suttas gives 3 descriptive formulas for the stages of discipleship (sāvakatta).20 They are that of “full discipleship,” that is, up to arhathood;21 that of the disciple teacher (one who is at least a streamwinner);22 and that of a path disciple (a streamwinner upwards). The last is listed in the (Aṭṭhaka) Puṇṇiya Sutta (A 8.82), thus:

(1) one has faith, saddho (2) approaches (the Tathagata), upasaṅkamitā

15 A 1:24; DA 1:27; MA 1:4, 2:247, 337, 3:27; SA 1:5, 2:94, 141; AA 1:5, 296; KhpA 101; UA 11; ItA 1:22, SnA 1:135; ThaA 3:112; ApA 128, 309. 16 A 1:27; AA 1:418 f, 445; DhA 1:210; ItA 1:29, 32; PmA 3;674. 17 S 2:235; AA 1:385. 18 M 65,32/1:445 (SD 56.2), 134,6/3:201 (SD 109.11); S 4.16/1:113 (SD 61.11), 8.6/1:189 (SD 58.7, qv), 46.38/- 5:95 (SD 41.10). 19 Odahati (+ odheti), “places (in); lays down; applies; directs, odahati sotaṁ, “gives ear attentively,” V 1:9,14). Cf oharati which also gives ohita but with a different sense (as in ohita,bhāra, “laid down the burden”). 20 Sāvakatta (sāvaka + -tta), the state of a true disciple: Upāli S (M 56/1:375 passim, 379, 380-382 passim), SD 27.1; (Licchavī) Sīha S (A 8.12/4:185), SD 71.5; Ap *1.13; V 1:236. 21 Caṅkī S (M 95,20), SD 21.15; Kīṭā.giri S (M 70,23 etc), SD 11.1. 22 (Dasaka) Puṇṇiya S (A 10.83,19), SD 57.23. http://dharmafarer.org 5 SD 58.1 The Oral Tradition of the Early Buddhists

(3) attends (on him), payirupāsitā (4) questions him, paripucchitā (5) listens to the Dharma attentively; ohita,soto dhammaṁ suṇāti (6) the Dharma, he holds it in mind; sutvā ca dhammaṁ dhāreti (7) of that Dharma that he remembers, he examines its teachings, and dhammānaṁ atthaṁ upaparikkhati (8) understands both the Dharma and its meaning, and attham aññāya dhammam aññāya practises in accordance with the Dharma,23 dhammânudhamma,paṭipanno hoti (A 8.82), SD 57.22 (1.2.1.1)

Even though the “8 stages of discipleship” refer to at least the attaining of streamwinning, they also apply to our commitment to the Dharma-spirited life that arises with (1) faith, that is, wise faith, our joy in hearing the Dharma and desire to learn it. Hence, (2) we approach the teacher, and respect him so (as a teacher): (3) we attend to him appropriately. Since a true teacher places the teaching above himself, we see this in him, and (4) question him on the Dharma. This is a living verbal exchange—a direct transmission of the Buddha Dharma from teacher to pupil—which takes effect as “an oral tradition” in Dharma when we (5) listen attentively to it, and (6) hold it in mind. Doing so, (7) we remember the Dharma, which we then examine diligently, so that (8) we understand its meaning, and practise it as taught by the Buddha. Practising in this manner, we begin to directly see into true reality: we reflect and understand the true nature of impermanence. Then, we are moved to aspire to streamwinning in this life itself. In due course, we reach the path and, on attaining streamwinning, we become true disciples (sāvaka) [1.2.1.4]. We are those who have truly listened and heard the Buddha Dharma, and benefitted from it.

1.2.1.4 Early Buddhism began and prospered as an oral tradition: the Buddha taught the Dharma for 45 years (449-404) BCE [1.1.1.1]. He teaches the fourfold assembly (catu parisā, literally, “(those) sitting around”)24 of monks, nuns, laymen and laywomen, to a group of listeners, to the group of 5 monks,25 even to a single person, such as the monk Māluṅkya,putta,26 the wanderer Vacchagotta,27 the young householder Sigāla,28 or to a large assembly of wanderers,29 and so on. The Buddha or his disciple teaches an audience that sits up close (m upāsaka, fem upāsikā), so that they may become true listeners (m sāvaka, fem sāvikā). The teacher, in such a living system, observes the pupil, and applying various skillful means, inspires him to live a moral life (sīla), not only to live harmon- iously, but as a basis for his mental cultivation (samādhi). When the pupil shows some level of maturity in both moral virtue and mental cultivation, he begins, with the teacher’s help, to grow in wisdom (paññā). Even for the wisest teacher, he could at best give the pupil only a theoretical presentation of the teaching; but if the pupil is ready, he would digest this presentation, even at once, see directly the true reality in it for himself. In such a case, the mature or awakened pupil would be accomplished in both knowledge and conduct (vijjā,caraṇa) [3.3], that is, wisdom with compassion, so that he is himself capable of using skillful means to uplift others.

23 Attha and dhamma are the 1st 2 components of the 4 analytic skills [5.4.2.13]. 24 On the fourfold assembly (catu,parisa), see Mahā,parinibbāna S (D 16,3.35), SD 9; Pāsādika S (D 29,12) + SD 40a.6 (2.2); D 51,12 (1.1.2.1). 25 Dhamma,cakka Pavattana S (S 56.11), SD 1.1. 26 Cūḷa Māluṅkya,putta S (M 63), SD 5.8. 27 (Ānanda) Vaccha,gotta S (S 44.10), SD 2.16(5). 28 Sigal’ovāda S (D 31), SD 4.1. 29 Udumbarikā Sīha,nāda S (D 25), SD 1.4.

6 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

1.2.2 Discipleship through seeing (dassana)

1.2.2.1 From the 8-step discipleship formula [1.2.1.3], it is clear that the early Buddhist “oral tradition” is not merely one of listening, but has the purpose and goal—both senses found in the Pali word attha, which also means “meaning, sense”—of “seeing,” both on the physical level and the spiritual level. The purpose of Buddhist training entails seeing (dassana) the teacher.30 The purpose of seeing the teacher, especially a noble disciple (ariya,sāvaka)—a streamwinner etc—is to hear the teaching so that we see the noble truths (ariya.saccāna,dassana).31 Note, too, that here dassana appears twice, the former with the sense of physical seeing (meeting and listening), and the latter, spiritual seeing (knowing and understanding). Meeting the Buddha or a Dharma teacher is not merely a social event, but more so a spiritual encounter for the arising of what is higher than either the teacher or the pupil, that is the Dharma. The teacher, by his good and true example, is the source and occasion for the Dharma, which benefits the pupil as listener.

1.2.2.2 The best known case of Dharma-spirited seeing and listening leading to the fruiting of the path is that of the wanderer Sāriputta’s meeting with the newly awakened elder Assaji. On Sāriputta’s eagerness to know the Dharma, Assaji utters this famous quatrain on causality:

Ye dhammā hetu-p,pabhavā Whatever states that arise from a cause, tesaṁ hetuṁ tathāgato āha their cause the Tathagata [thus-come] has told, tesañ ca yo nirodho and their ending, too— evaṁ,vādī mahā,samaṇo thus spoke the great recluse. (Mv 1.23.5+10)32

This is another, slightly longer, version of the Buddha’s brief statement on “only suffering and the ending of suffering do I declare.”33 “Only suffering” (the 1st 2 truths) is reflected in the first 2 lines; and “ending of suffering” (the last 2 truths), in the last 2 lines. On hearing the 1st 2 lines, Sāriputta at once understands their meaning, and attains streamwinning. When he repeats this quatrain to his companion, Moggallāna, he, too, attains streamwinning. They both know that they need to learn directly from the Buddha to reach the path’s goal. In due course, they become the Buddha’s 2 chief disciples.34 Sāriputta’s transformation on account of seeing and listening to the arhat Assaji is called the Assaji effect. Sāriputta, by his own goodness, is inspired by seeing Assaji’s calm demeanour and self-restraint: this is a vision of a noble disciple. On hearing Assaji’s teaching, albeit a brief quatrain, Sāriputta’s attains the Dharma-eye and reaches the path: this is a vision of the noble truth.

1.2.3 DISCIPLESHIP AS VINAYA

1.2.3.1 The early Buddhist oral tradition (mukha,pāṭha) comprises not only the Dharma, the Bud- dha’s teaching of true reality, but also the Vinaya, the monastic rules and right livelihood. While the Dharma embodies what the Buddha himself discovers and teaches so that we can follow the same path

30 On “seeing the teacher,” esp the Buddha, see Kāranapālī S (A 5.194), SD 45.11. On “not seeing holy ones,” M 135,17 f (SD 4.15). On “seeing the Buddha,” SD 52.1 (II). 31 On ariya.saccāna,dassana, see Khp 5.10 = Sn 267 (SnA 151,24). On dassana: SD 56.22 (8.4.2.2). 32 Mv 1.23.5+10 @ V 1:40,28*+41,35* (VA 5:975) = Ap 1.146/1:25 (ApA 231) = Peṭk 10 = ThaA 3:95. See also SD 49.20 (1.2.2.2); SD 56.4 (3.6.3.2). 33 Anurādha S (S 22.86), SD 21.13; Alagaddûpama S (M 22,38), SD 3.13; SD 40a.1 (11.1.1). 34 See Piya Tan, The Buddha and His Disciples, [2002] 2013 ch 5. On their decision to renounce during the hill-top festival, see SD 52.2f (1.4). http://dharmafarer.org 7 SD 58.1 The Oral Tradition of the Early Buddhists of awakening, the Vinaya are the social conditions for preserving the teachings and practice rooted in the spirit of renunciation (nekkhamma). The Vinaya provides the right conditions for those wishing to renounce the world and worldliness to gain the path of awakening in this life itself. It is introduced for the benefit of those who are not yet on the path of awakening so that they can attain it here and now. Monastics, by their own choice, join the order as renunciants to follow the Vinaya so that they are properly supported by the faith of the laity.

1.2.3.2 The Vinaya also protects the laity from any wrongdoings of monastics, who being unawakened are still driven by their defilements. Many of the rules have been introduced on account of complaints or requests from the laity. Hence, the Vinaya is also a kind of “social contract” between the monastics and the laity. Lay followers are thus reminded not to distract the monastic in any way from their avowed life of renunciation dedicated to working for the path, but to be a support to them in that effort here and now. In this way, the laity, too, is reminded of the discipleship of the monastics and that of the laity. Monastics who disregard or reject the Vinaya—especially the Pāṭimokkha (monastic code) and its fortnightly conclave—have failed in their monastic life and renunciation. When we respect or support such monastics, we are, in effect, supporting and encouraging such a failure and wrongdoing. There is no sangha for monastics who do not uphold the Vinaya as handed down from the Buddha’s time.

1.2.3.3 On the other hand, as the laity, we can and should follow the Buddha Dhamma, without such false monastics. Moreover, we have taken the 3 refuges: the Buddha as our one and only true teacher; the Dharma as the teaching that leads us to the path; and the noble sangha (of monastic and lay streamwinners, once-returners, non-returners and arhats) who are our exemplars. This Dharma continues in our personal practice even when worldly conditions may not be favourable to the Dharma-Vinaya.

2 Living the Dharma

2.1 There are ancient Pali texts that record teachings by monastics and by lay followers; some by Brah- ma or Sakra; some by non-human beings; and some are traditional statements, which, due to their significance, were interpolated into the Canon after the Buddha’s time. Most suttas, of course, record conversations between the Buddha and such audiences. From the way a sutta is formally structured and worded [3], often with more than merely the Bud- dha as speaker or narrator, it is clear that such texts have been compiled and edited in some way—the way we would edit a recorded talk into a readable book today. However, from the common textual or literary conventions and the universal unity of ideas—many of which clearly explain one another—these texts are not a haphazard or random composition. They have been put together with unity of narrative and lesson [4.2.1]—they especially inform and instruct, almost following the template of a computer programme. The longer suttas, unified and classic in their structure, may tell a story, or may serve as a careful guide to some teaching or some meditation practice. As a rule, such teaching or practice can only be properly effected by reading the text faithfully with its repetitive passages [4.2.2]. These recurrent passages are not merely mnemonic devices but they also serve as vital self-inductive lessons in moral virtue, mental cultivation or insight wisdom: they need to be read in full and faithfully—to live the complete sutta. [7.2]

2.2 Some of the Pali texts that we now have are very old, clearly going back to the Buddha’s life-time. Clear examples are those of the Aṭṭhaka,vagga (the Chapter of Eights) and the Pārāyana,vagga (the Chapter of Going to the Far Shore), which once existed as independent texts. With canonization of the

8 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists texts, they are included in the Sutta Nipāta of the Pali Nikāyas, and are also found in the Sanskrit Āgamas (the Saṁyukta Āgama and the Ekôttara Āgama). We now know that many sutras––in Pali, Sanskrit or any other Indic language—are most likely not spoken verbatim by the Buddha himself, but their content and drift are clearly those of early Buddhism. This is clearly not the case with the Mahayana texts, composed much later, containing the same passages, phrases or words purported to have been spoken by the Buddha. To the extent that such passages have not been “revised” by Mahāyāna ideas, it is possible to accept them as Sakyamuni’s genuine teachings insofar as they embody the true spirit of Buddha-word.35 [3.2]

2.3 It is almost impossible for us to be sure of the exact texts or words of what the Buddha said or taught because we have no ancient manuscripts or records of any kind going back to the Buddha’s time. However, we do have what may be called the “mind-transmission” of his teachings handed down by the contemplative monks, especially the forest monks. Such monastic practitioners, who faithfully uphold the early Dharma-Vinaya, preserve the Buddha’s meditation teachings and experiences in their person and practice. 36 The Buddha’s teaching can never be fully understood, much less, its goal attained, without our experiencing some level of meditative stillness. Like erudite and insightful scholars, we may know what the Buddha taught, but we can never know what he meant or realized—without the full experience of the kind of meditation that he himself had used. What we know from the suttas, and what we see from such meditative insights, can only be authenticated by the clear vision of our own meditative minds. The Bud- dha Dharma is, in other words, a living transmission.

3 Reading and listening

3.1 SUTTA STUDY PROMOTES MINDFULNESS What if we do not have a high level of meditation ability, or even none at all? Would we be able to understand what the Buddha teaches? Without some experience of proper meditation, the Buddhist teachings are at best literature. Indeed, even as literature, the Buddhist canon is a huge collection, many times larger than those scripture of the book religions.37 A good way to truly appreciate the suttas on a spiritual level, is to attend teachings given by experienced Dharma teachers. As our understanding of the suttas deepen, we become more inspired and empowered toward meditation practice.

3.2 IMPORTANCE OF THE NIKĀYAS

3.2.1 The early Buddhist texts (EBTs)

3.2.1.1 By “early Buddhist texts” (EBTs) are meant the suttas and teachings, as we have them today, have been spoken or given by the historical Buddha and his immediate disciples or based on such teachings. By “authenticity” is meant that these texts and teachings were actually taught or spoken by the Buddha or his immediate disciples. Just as we would today record the teachings of an experienced Dharma teacher, and then edit or redact it into articles, booklet, books and sets, especially convenient for a reading public, the teachings of the Buddha and his immediate disciples have been redacted and edited for the oral tradition as a means of dissemination to the students, monastic and lay. It is vital to bear this in mind so that we fully and truly

35 See Joy Manné 1990, 1992. 36 Assaji is a good example of a monk who protects or embodies the Dharma in his body [1.2.2.2]. 37 The Pali Canon itself is estimated to be over 11 times the Christian Bible in size. http://dharmafarer.org 9 SD 58.1 The Oral Tradition of the Early Buddhists appreciate the early Buddhist texts and not present a Procrustean truncated Dharma: that would be mis- representing the Buddha.38 While it was true that the monastic community were more right and ready as reciters and preservers of the oral tradition of the early Buddhist texts, a significantly large number of lay disciples—who like the monastic disciples, were also noble saints (ariya)—were also proficient in the teachings, and were preservers and disseminators of the Dharma. The lady Khujj’uttarā [4.2.4.1], for example, committed to memory teachings of the Buddha that has been canonized as the Iti,vuttaka.[5.4.2.6]

3.2.1.2 We have already noted that the early Buddhist teachings were authentic and orally transmitted [4]. This is clearly evident from its canonical texts, especially the suttas of the first 4 Nikāyas, the older books of the 5th Nikāya and some Vinaya texts, constituting the early Buddhist texts, that is: the Dīgha Nikāya the collection of long discourses; the Majjhima Nikāya the collection of the middle-length discourses; the Saṁyutta Nikāya the collection of the connected discourses; the Aṅguttara Nikāya the collection of the numerical discourses; the Khuddaka Nikāya the small collection (the “5th” Nikāya), especially the Sutta,nipāta, the Dham- mapada, the Udāna, the Iti,vuttaka, the Thera,gātha and the Therī,gāthā; the Vinaya the Patimokkhas (pātimokkha); some materials from the Khandhakas.39

Buddhist texts that are not “early” (spoken neither by the Buddha nor his immediate disciples) are most of the Buddha biographies, most of the Khuddaka Nikāya, and most of the Vinaya Piṭaka. The “not- early Buddhist texts” (NEBTs) also include the Abhidhamma, the Abhidharma and the Mahāyāna Sūtras. As for the Jātakas, only the verses are early (canonized as the Jātaka), but the very same word, Jātaka, often also refer to the stories themselves, which are not early texts. The Commentaries and other late texts (subcommentaries, compendia, etc)—including the traditional chronicles (vaṁsa)—may contain some genuine historical material alongside mythical accounts and much later compositions. Some of these texts have been written as recently as a few centuries ago. De- spite their lateness, where they present materials based on the early texts, they may serve as helpful “extratextual” supportive readings of the early texts, such as the retelling of the Buddha-story.40

3.2.2 The 4 Nikāyas

3.2.2.1 A special characteristic of these early Buddhist texts is that they tend to be very repetitive or cyclic and often relate to one another. They are to be read or heard, and reflected on: this is the start of our proper Dharma practice. This vital characteristic is also a mnemonic tool that helps reciters remember the teachings so that they are properly preserved right down to our own time as the Pali canon, the oldest parts of which are the 4 Nikāyas. Each of the 4 Nikāyas has its special textual characteristic, besides being “long,” “middle-length,” “connected” and “numerical,” respectively. The Dīgha Nikāya, for example, presents the “long” discourses, dialogues and debates of the Buddha in a broad and dramatic manner mostly to attract non-Buddhists. These often elaborate suttas are skill- fully presented, as it were, like a prospectus for a broad-based course.

38 See Anumāna S (M 15,7.3 n) + SD 59.3 (1.1.2) n. 39 These early texts preserve the ancient records of the ordination (upasampadā) and the observance (uposatha) procedure, that are found across all Vinaya traditions. See E Frauwallner, The Earliest Vinaya and the Beginnings of Buddhist Literature, 1956:78 f. 40 See Sujato & Brahmali 2015:6-10.

10 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

The Majjhima or “middle-length” collection comprises further dialogues of the Buddha, some of them are longer than those of the Dīgha. The Majjhima suttas are intended to inspire the monastics and the faithful laity to appreciate the spirit of the Dharma. The Saṁyutta Nikāya arranges the Buddha’s teachings into “connected” chapters and sets, that is, there is less narrative but more instructions here. Like the Majjhima, the Saṁyutta is rich in instructive materials, but generally more episodic and much shorter than those in the Majjhima. The Aṅguttara Nikāya arranges sutta materials in a “numerical” manner, which would be easier for the laity (for whom it is clearly intended) to appreciate, so that they are inspired towards more committed Dharma practice. However, in its later sections, it has comparatively longer suttas, like those in the preceding 2 Nikāyas. [6.2.4 The 5 Nikāyas]

3.3 THE SUTTAS IN PROGRAMME LANGUAGE

3.3.1 Listening tools There is another interesting characteristic of the suttas, especially those of the 4 ancient Nikāyas, not yet fully explored by any academic or specialist in our times. The sutta materials are not merely repetitive, but they are patterned like a computer programme.41 For example, there are recursions or recursive sequences (where a template, say, lists the 5 aggregates, and each template describes or defines one aggregate at a time) [3.3.2.1]; concatenation or connected series (where the template or passages mention an additional word or expression each time incrementally) [3.3.2.2]; and nesting or “concentric” loops (where a teaching or narrative is contained within another on a number of layers). [3.3.2]

3.3.2 Examples of programming mechanism

3.3.2.1 The suttas are replete with recursions, concatenations, nestings and other “programming” devices in its template-like format to help us avoid “intellectualizing” what we read or hear, assuming that we know or understand it, and so not learning anything useful. These sutta devices help us “listen” to the teaching just as it is taught by the Buddha by way of helping us see it as a whole and remember it better for personal practice.

3.3.2.2 Here are 3 examples of recursion in the suttas: (1) the full recursive cycle in the Cha Chakka Sutta (M 148), (2) the recursive cycle in the definition of the 5 aggregates in the Anatta Lakkhaṇa Sutta (S 22.56), and (3) recursion in the definition of priorities in the Ādhipateyya Sutta (A 3.40).

(1) The recursive cycle is fully applied in the Cha Chakka Sutta (M 148) to the formula for “the way leading to the ending of self-identity” towards the 6 internal sense-bases, the 6 external sense-bases, the 6 groups of consciousness, the 6 groups of contacts, the 6 groups of feelings, and the 6 groups of cravings (these are the “6 sixes”), by way of the following 3 pairs of teaching cycles:42

(1) §§4-9, the syllabus, and §§10-15, the exposition on non-self; (2) §§16-21, the arising of self-identity, and §§22-27, the ending of self-identity; (3) §§28-33, latent tendencies, and §§34-39, abandoning latent tendencies.

(2) Another recursive pattern is seen in the definition of the 5 aggregates in the Anatta Lakkhaṇa Sutta (S 22.56), thus:

41 See SD 50.16 (1.1.2). 42 M 148/3:280-287 (SD 26.6). http://dharmafarer.org 11 SD 58.1 The Oral Tradition of the Early Buddhists

3 (1) “Bhikshus, form is non-self. For, bhikshus, if form were self, this form would not bring about illness [affliction], and it would be possible to tell the form: ‘Let my form be such. Let my form not be such.’

4 But, bhikshus, because form is non-self, form brings about illness [affliction], and it is not possible to say of form: ‘Let my form be such. Let my form not be such.’ Then, the same pattern is successively used for each of the other 4 aggregates—feeling, perception, formations and consciousness—in its turn.43

(3) We again see basic recursion pattern working in the definition of priorities (ādhipateyya) in the Ādhipateyya Sutta (A 3.40). First, each of the 3 priorities—self-priority, world-priority, Dharma-priority —is mentioned. The subject is the renunciant, who reflects that he has not renounced for the sake of any of the worldly requisites: food, robes, or lodging, or even comfort (usually, this is health). Then, in the 1st cycle, reflecting on his own renunciation, he determines that it is for “his own purity,” his spiritual growth. In the 2nd cycle, reflecting on the world-community, filled with the wise who knows his actions. And in the 3rd cycle, reflecting on the Buddha’s well taught Dharma, and his fellow brahma- charis (colleagues in the holy life), he commits himself to his practice.44

3.3.2.3 A concatenation is a connected series where the template or passages mention an additional word or expression each time incrementally. We see this feature at work in, for example, the description of the Buddha’s disciples in Mahā Vaccha,gotta S (M 73), which, incidentally, also uses the recursive sequence. The descriptive listing progresses as follows: the monastic arhats (monk arhats and nun arhats), the laymen saints (non-returners, once-returners, streamwinners), and the laywomen saints (non-returners etc). Each of these categories of noble individuals are each said to be in great numbers.45

3.3.2.4 Nesting or “concentric” loops contain teachings or stories within another on a number of levels or layers. The Sigāl’ovāda Sutta (D 31), for example, opens with a teaching on the 6 directions (D 31,1-2); then, it breaks off in nested cycles of teaching (the 4 defilements of conduct, the 4 motives, the 6 ways of squandering wealth, and the 4 false friends and 4 true friends) (D 31,3-26); and climaxes with the teaching on the “6 directions” (D 31,27-35).46 Nesting is more commonly found in stories. The Sāmā,vatī Vatthu (the story of Sāmāvatī) is a classic example of multiple nested stories, covering past lives, too (DhA 2.1). Its cycles of episodes are as follows:47

(1) Udena’s birth and youth. (2) Ghosaka’s birth and youth. Story of the past: Kotūhalaka casts away his son. Story of the present: Ghosaka is cast away 7 times. (3) Sāmāvatī’s birth and youth: (4) Udena wins Vasulā,dattā: Story of the past: Caṇḍa,pajjota wins the 5 vehicles. (5) Buddha’s rejection of Māgandiyā. (6) Deaths of Sāmāvatī and Māgandiyā: Treasurers, monks and tree-spirits; story of the past. Khujj’uttarā converts Sāmāvatī.

43 S 22.56,3-11/3:66 f (SD 1.2). 44 A 3.40/1:147-149, prose section (SD 27.3). 45 M 73,7-12/1:490 f (SD 27.4). 46 D 31/3:180-193 (SD 4.1). 47 Better known as the “king Udena story-cycle” (DhA 2.1/1:160-228); tr in Burlingame’s Buddhist Legends (DhA:- B 1:247-293.

12 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

Māgandiyā’s plot against Sāmāvatī and the Buddha. Burning of Sāmāvatī and Māgandiyā’s punishment. Story of the past: Sāmāvatī’s past karma. Story of the past: Khujj’uttarā’s past karma. 3.3.3 Dharma in fractals

3.3.3.1 The whole of the Pali canon, the collection of early Buddhist texts, is patterned on the 4 noble truths (catu ariya,sacca) from the smallest point to the broadest collection of teachings. On the most basic word level, we have the terms, dukkha, samudaya, nirodha, magga or dukkha, samudaya, magga, nirodha.48 Each of these terms may be expanded indefinitely, encompassing a growing amount of details as needed.49 On the broadest level, as a complete collection of texts, the 4 noble truths form the basic structure of the Saṁyutta Nikāya, the connected collection. According to the parable of the elephant’s hoof-print, the footprint of any living being can be placed in that of an elephant’s hoof (on account of its foremost size). Even so, “all wholesome states are included or found in the 4 noble truths.”50 This is not modernist fad, but merely a contemporary vision of the Buddha’s declaration: “Only suffering and the ending of suffering do I declare.”51

3.3.3.2 A fractal is a mathematical set that generates a “self-similar” object or pattern, one that mir- rors itself, like that in a snowflake, that is, exactly or approximately similar at all scales. Unlike normal geometric patterns, fractals retain the tiniest details at any level. “Fractal,” coined by Polish-French math- ematician Benoit B Mandelbrot (1924-2010), comes from Latin fractus, meaning “broken, fragmented,” in the sense of proliferating into numerous identical “self-similar” components. Such patterns were used in pre-modern art even before science discovered fractals. 52 The early Buddhist texts have fractal characteristics both in content (the words), form (the language) and style, such as the common use of repetition. As already noted [3.3.1], these repetitions recur as recursions, concatenation, nesting and so on, at the level of word, sentence, paragraph, passage, sutta and the group of suttas, such as the Saṁyutta Nikāya [3.3.3.1].

3.3.3.3 That a sutta teaching resembles fractals is not coincidental. A fractal occurs by the recursive iteration of a simple operation, as does a Dharma teaching in detail, especially through repeated questioning (paṭipucchā) of the previous statement: “And what, avuso (or bhikshus), is … ?” as we see in the Mahā Hatthi,padôpama Sutta (M 28).53 Just as the fractal patterns are compact and compressible, it is possible for us to learn the essence of the Dharma in a few basic words, and all expansions exemplify the same pattern. Such patterns are frequently observed in nature as a key feature of natural growth as opposed to artificial creation.

48 The former is the “teaching model” of suffering, arising, ending and path; the latter is the “practice model”: see SD 1.1 (6.2.2.2). 49 See eg Mahā Hatthi,padôpama S (M 28, SD 6.16) + MĀ 30; Sacca Vibhaṅga S (M 141, SD 11.11) + MĀ 131 / T32 / EĀ 27.1 50 M 28/1:184-191 (SD 6.16). 51 See Anurādha S (S 22.86), SD 21.13; Alagaddûpama S (M 22,38), SD 3.13; SD 40a.1 (11.1.1); SD 58.1 (1.2.2.2). 52 Gale Ency of Science, 4th ed, 2008:1809-1811, fractal. Oxford Dict of Science, 6th ed, 2010: fractal. Scott Thornbury, F is for Fractal. 29 April 2012, https://scottthornbury.wordpress.com/2012/04/29/f-is-for-fractal/. Eric W Weisstein, “Fractal.” From MathWorld—A Wolfram Web Resource. 2013, https://mathworld.wolfram.com/Fractal.html. 53 M 28,2/1:184 f (SD 6.16). http://dharmafarer.org 13 SD 58.1 The Oral Tradition of the Early Buddhists

The fractal features of the EBTs, therefore, suggest that they also grew organically and hol- istically, not through the artificial bolting together of elements from different origins. The most obvious way that this would have happened is through a gradual expansion in details and contexts of the earliest teachings. The initial impetus for teaching may have been a particular insight, but an insight that allowed virtually endless variations and expansions in its exposition. This differs from non-EBT Buddhist literature, where we frequently see early elements juxta- posed with entirely different formulations. (Sujato & Brahmali, The Authenticity of the Early Buddhist Texts, 2005:99)

4 Oral tradition as literature

4.1 THE EARLIEST BUDDHIST TEXTS

4.1.1 Meaning of “text”

4.1.1.1 The word “pāli” (also spelt pāḷī) is the proper term for a sutta, whether in part or in full, or even a passage, that is narrative or didactic—whether it recounts stories related to teachings or explains the teachings themselves. Technically, the contents of all the 5 Nikāyas, the Vinaya and the Abhidhamma is called pāli, which then simply means “text.” Outside of the pāli or canon of texts, are the Commentaries, aṭṭhakathā (or atthakathā). “the talk on meanings.” The term “commentary” would then include those texts that are not canonical, arose after them. Provisionally, to keep this taxonomy simple, we can include all other late texts, such as the guides to canonical teachings, the traditional chronicles, Abhidhamma-related works, grammatical works, hagio- graphies (legends of the Buddha, the buddhas and saints), and so on.54

4.1.1.2 There are a few anomalies we must be aware of, especially when we are beginning a serious study of the early Buddhist texts [3.2.1]. Firstly, the Niddesa—comprising Mahā Niddesa (Nm) and the Cūḷa Niddesa (Nc)—are a pair of commentarial work, but they are so ancient—early Buddhist texts—that they have been included as the 11th book of the Khuddaka Nikāya. The Mahā Niddesa (or Niddesa I) is the canonical commentary on the Aṭṭhaka,vagga (An 766-975); and the Cūḷa Niddesa (or Niddesa II) is the canonical commentary on the Pārāyaṇa,vagga (Sn 976-1149, end of Sn) and the Khagga,visāṇa Sutta (Sn 35-75). The Niddesa, despite being a commentary—hence, not a work of the oral tradition—was probably included in the Pali canon not earlier than Asoka’s time, after which the canon was closed.55 [5.4.2.1(3)]

4.1.1.3 The term “paracanonical” is alien to Pali, but it is a useful literary term for our usage and facility. Simply, a paracanonical text is not included in the Pali canon, but it is not a commentarial work. The term “paracanonical texts” is used by Oskar von Hinüber (1996:76). K R Norman uses the term “early post-canonical texts” (1983b:108). These are ancient texts (commentarial works) which explain various difficult points in the suttas. There are 4 well known paracanonical Pali works.56 The oldest of these seem to be the Peṭakopadesa (the Pitaka disclosure; Peṭk), which seems to be an earlier draft of the better written Netti-p,pakaraṇa

54 For a more detailed analysis on this taxonomy, see eg K R Norman, Pali Literature, 1983b; Oskar von Hinüber, A handbook of Pali Literature, 1996. 55 It should be further noted that the Sutta Nipāta (Sn) has another Commentary, Paramatthajotika II (attr to Buddhaghosa) and abbreviated as SnA (which is different from NmA and NcA, which are Comys to Mahā Niddesa and Cūḷa Niddesa respetively). See Norman 1983b:84-87 (see index: Niddesa); Hinüber 1996:58 f. 56 See Norman 1983b:108-114; Hinüber 1996: Peṭk 80-82, Nett 77-80, Miln 82-86, Vimm + Vism 123-126.

14 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

(the guide; Nett), which is a exegetical guide on early Buddhist hermeneutics, that is, a guide for those who write commentaries. They are very technical works that directly quote the suttas. Little is known of the actual of authorship or exact dates of their composition. While the Peṭakôpadesa was probably written no earlier than the 2nd century BCE, the Netti-p,pakaraṇa, around the beginning of the Common Era. The next well known or “classic” paracanonical text is the Milinda,pañha (Milinda’s questions; Miln), an early work on Buddhist apologetics, originally written in Sanskrit or Prakrit in north India around the beginning of the Common Era, and translated into Pali. Understandably, the answers given by the monk Nāgasena to the questioner, Bactrian Greek king Milinda (Menander), occasionally do not fully reflect the spirit of the early teachings. The last of the classic paracanonical text is the Visuddhi,magga (the path of purity; Vism), written by the best known of Pali commentators, Buddhaghosa (from north India) around the 5th century CE. Dis- counting Buddhaghosa’s personal views and those of the Mahāvihāra (the dominant Theravāda centre under whose auspices he wrote the classic), the work is a helpful introduction (prolegomenon) to the 3 trainings [1.2.1.1], that is, the entire Pali canon. The commentator Dhammapāla, a South Indian commentator, who lived just after Buddhaghosa, wrote the Param’attha,mañjusā (VismṬ), a commentary on Vism. Dhammapāla tells us that the word ekacce, “some or a certain (authority),”57 refers to the elder Upatissa, author of the Vimutti,magga (the path of freedom; Vimm), composed just before the 5th century in north India, and was fully translated into Chinese in 505. Its original is lost and today we only have the Chinese version. It is well known that Buddhaghosa was familiar with it, and used it as the basis for his own Visuddhi,- magga, but without mentioning it at all. The reason is probably because Vimm belonged to the rival Abhaya,giri tradition, while Buddhaghosa was writing for the Mahāvihāra.58

4.1.2 Evidence for oral transmission in early Buddhism

4.1.2.1 According to Indologist, Mark Allon (Style and Function, 1997), it is generally agreed that early Buddhist literature, of which the Pali texts of the Theravāda canon are the most numerous and best preserved examples, was composed and transmitted orally.59 Allon gives the following reasons in his paper, “The oral composition and transmission of early Buddhist texts” (1997a):

(1) There is no reference to writing or writing materials in the principal Pali Nikāyas60 though there are many references to learning and reciting suttas.61 (2) Although there are a few passages in the Vinaya Piṭaka which indicate that the art of writing was known at the time when these Vinaya texts were put into their present form, these do not refer to the early texts or their preservation.62 (3) Despite detailed rules governing the use of all items that are used by monks and nuns, the Vinaya has no rules governing the use of writing materials.63 (4) There is no archaeological evidence for the use of writing in India during the early phase of Buddh- ism, that is, before the time of Aśoka64—although this view may have to be revised in light of the findings in Sri Lanka of Brāhmī characters on potsherds dating from this period. 65

57 Vism 3.80/102,31; see Vism:Ñ 104 n19; Norman 1983b:113 f§. 58 Hinuber 1996:123-126.. 59 0 von Hinüber 1990b: ch V (esp p22), p30, ch XIV; K R Norman 1993a:280; R Gombrich 1990a & 1990b; L S Cou- sins 1983 esp p1; S Collins 1992. 60 R Gombrich 1990b:27; 0 von Hinüber 1990b esp 30. 61 S Collins 1992:124 f; R Gombrich 1990b:26. 62 T W Rhys Davids & H Oldenberg 1881:xxxii-xxxv; cf R Gombrich 1990b:27 f. 63 T W Rhys Davids & H Oldenberg 1881:xxxii f; cf R Gombrich 1990b:28. http://dharmafarer.org 15 SD 58.1 The Oral Tradition of the Early Buddhists

(5) Finally, many of the stylistic features of these texts indicate an oral origin.66 (1997:39-61)

4.1.2.2 The suttas often depict monks and nuns learning and reciting the Dharma, which seem to indicate that even during the Buddha’s life-time, teachings were formulated for memorizing and reciting. It is reported in the Udāna and the Vinaya, for example, that the Buddha asks the monk Soṇa to teach the Dharma. In response, Soṇa recites the Aṭṭhaka Vagga (sabbān’eva aṭṭhaka,vaggikāni sārena abhāsi), an ancient group of suttas in verses, the name by which we now know them, preserved in the Sutta Nipāta.67

4.1.2.3 The Vimutt’āyatana Sutta (A 5.26) lists 5 ways in which a Dharma practitioner may attain spiritual freedom (vimutti), that is, (1) through “deep listening” (experiencing the Dharma), (2) by teaching it, (3) by reciting it, (4) by reflecting on it, and (5) through mental concentration.68 Of these 5 methods or occasions for spiritual freedom, the first 3 are directly related to the oral/aural tradition, that is, by way of teaching and listening. In the 1st case, it is the Buddha himself or a Dharma teacher who teaches the Dharma, and our “listening deeply” (paṭisaṁvedeti, “experiencing”) to it; in the 2nd, the monk (or any of us who are right and ready) teaches it (deseti); and the 3rd is that of our reciting the Dharma ourself (sajjhāyaṁ karoti). The key points of this teaching are listed in the Saṅgīti Sutta (D 33).69

4.1.2.4 The Vinaya records the occasions when monks who are sutta-specialists reciting a sutta in groups (suttantikehi suttantaṁ saṅgāyantehi).70 There is also an especially interesting pācittiya rule that prohibits monastics from teaching the “unordained” (a non-monastic) the Dharma by making them (the unordained) recite it word by word or “line by line” (padaso dhammaṁ vāceyya).71 This rule was made by the Buddha because certain monks were teaching some laymen in this manner. The Vinaya’s Old Commentary (an embedded autocommentary) takes this as a particular form of recitation, probably in the manner of Vedic chanting. Buddhaghosa, in his Vinaya Commentary, explains this as letter by letter (anvakkharaṁ), syllable by syllable (anuvyañjanaṁ), word by word (pada), line by line (anupada), verse by verse (gāthā) and section by section (koṭṭhāsa) (VA 741,8-23). This hints at the Vedic recitation style used for apprentices, which is not encouraged in Buddhist training [4.1.2.4]. Any- way, the point remains that students were taught to recite and memorize certain texts.

4.1.2.5 Further, a definitive part of the monastic life is the fortnightly conclave and recitation of the Pātimokkha, the basic code of monastic rules. On one occasion, for example, the Buddha refuses to recite the Pātimokkha because the assembly of monks is “not pure,” meaning that one of the monks present has broken a rule and not make amends for it.72

64 R Gombrich 1990b:27. On writing in India, see 0 von Hinuber, 1990, esp 54, 72, and K R Norman 1992 [1993h] esp 243, 245-247. 65 See R A E Coningham 1993. S U Deraniyagala 1990 dates these finds to 600-500 BCE; R Allchin & R A E Coning- ham 1995 tentatively date them to 400-450 BCE. Cf F R Allchin 1989. 66 0 von Hinüber 1990:22 f; R Gombrich 1990a:7 f, 1990b:21 f; L S Cousins 1983; G von Simson 1965; 1977:479. 67 V 1:196 f. U 59 reads soḷasa aṭṭhaka,vaggikāni sabbān’eva sarena abhaṇi. Aṭṭhaka Vg (Sn 766-975), together with the Pārāyana Vg (Sn 976-11249) believed to be the oldest texts we have of the Pali canon: SD 49.6 (1). See S Collins 1992:125. 68 A 5.26/21.5 (SD 21.5). 69 D 33,2.1(25)/3:241 f; cf Collins 1992:126 f. 70 Mv 4.15.4 (V 1:169). 71 V 4:14 f; PED sv pada: padaso. 72 Cv 9.1.1-4 @ V 2:236-241 = U 5.5/51-56 (SD 59.2b).

16 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

Finally, a well-trained disciple, is often defined as one who is “learned” (suta, “heard”),73 who has heard much (bahu-s,suta), mastered attainment [5.2.1.2], Dharma-expert, Vinaya-expert, Matrix-expert,74 and so on.75 These are not titles for what we today know as “professional” or “qualified” experts, but descriptive names for monastic practitioners and disciples who have dedicated themselves to studying and mastering those aspects of the Buddha’s teaching not only for their spiritual growth, but also for the benefit of others.

4.2 THE SOUND NATURE OF THE EARLY ORAL TRADITION

4.2.1 Narrative or didactic?

4.2.1.1 Scholars have 2 main theories for the oral composition and transmission of the early Buddh- ist texts. We will briefly examine the 1st theory here, and the 2nd later [4.2.1.5]. The 1st theory is based on that proposed by Milman Parry and Alfred B Lord for oral-formulaic Hom- eric epic literature in Yugoslavia, and taken by Lord as the only method possible in oral cultures. In such an improvisatory method no two performances are identical.76 If the early phase of Buddhist literature was one of impromptu oral composition—texts were put together as needed depending on the occasion—then, those texts which we consider to be representative of this period must be seen to be “frozen” versions of a particular performance.77

4.2.1.2 Some scholars have emphasised the role of improvisation, and argued that the early Buddh- ist texts were modified and adapted according to the particular conditions of performance. This theory has been applied to the early Buddhist texts by L S Cousins as follows:

In practice they would have to be tailored to the needs of the particular situation—shortened or lengthened as required. An experienced chanter would be able to string together many different traditional episodes and teachings so as to form a coherent, profound and moving composition. It has been clearly shown that in many cases a traditional oral singer does not have a fixed text for a particular song. He can for example be recorded on two different occasions. The result may vary in length. (L S Cousins, “Pali oral literature,” 1983:1)

Cousins argued that in the earliest phase in the production of Buddhist texts, the monks performed accounts of the Buddha’s teachings and presented them in the manner proposed by the Parry-Lord model, that is, with “a strong improvisatory element” (1983:9). Over time, this material came to be fixed due to its religious authority (6).

73 Our modern idiom for this would be “well-read.” 74 Comy glosses māṭikā,dhara as “experts in the 2 matrices” (dve,māṭikā) (AA 2:189, 3:382), which Subcomy says are either the 2 Pātimokkhas (of the monks and of the nuns), or the summaries of the Vinaya and the Abhidhamma (AAṬ:Be 2:89). The Abhidhamma matrix is a list of keywords for mental states that in post-Buddha times grew into the Abhidhamma tradition: see Gethin, “The Mātikās,” 1992:156-164, esp 161 f for def of mātikā,dhara. See 5.3.2.2. 75 Eg, Sekha S (M 53): bahu-s,suto hoti suta,dharo suta,sannicayo (M 53,14/1:356), SD 21.14; so ca bhikkhu bahu- s,suto hoti āgatâgamo dhamma,dharo vinaya,dharo mātikā,dharo paṇḍito vyatto medhāvī lajjī kukkuccako sikkhā,- kāmo (Mv 2.21.2 @ V 1:119,22-24 = Mv 10.1.2 @ V 1:337,12-14, V 2:8,28); bhikkhu bahu-s,suto hoti āgatâgamo dhamma,dharo vinaya,dharo mātikā,dharo: D 16,4.10/2:125,6+25 (SD 9), A 3.20/1:117,28 (SD 37.2b), 2:147,31; V passim). Cf S Collins 1992 (for a brief description of teaching and learning process, see esp p124). 76 See Albert B Lord, The Singer of Tales, 1960. 77 See L S Cousins 1983 & R Gethin 1992. http://dharmafarer.org 17 SD 58.1 The Oral Tradition of the Early Buddhists

4.2.1.3 Against this theory, others have argued that the early Buddhist literature is different in many ways from the sort of oral material that is formed in performance. K S Norman has pointed out the following:

The great majority of Pali canonical texts, however, are in prose, and complete accuracy of reproduction is required at each recitation. In these circumstances the findings of modern inves- tigators of oral epic literature seem to have little relevance. (Norman, A Philological Approach to Buddhism, 1997:49)

Furthermore, R F Gombrich has pointed out that the peculiar nature of the early Buddhist texts makes it more likely that precise wording mattered in its transmission. In a paper entitled “How Mahā- yāna began” (1990b), Gombrich argues against the improvisatory stage proposed by Cousins:

The early Buddhists wished to preserve the words of their great teacher, texts very different in character from the general run of oral literature, for they presented logical and sometimes complex arguments. The precise wording mattered. (Gombrich, “How the Mahāyāna began,” 1990b:21)

He sees the early Buddhist texts as “deliberate compositions which were then committed to memory, and later systematically transmitted to pupils” (24), because, he adds, “the whole purpose of the enterprise ... was to preserve the Buddha’s words” (22).

4.2.1.4 If, according to Gombrich, verbatim accuracy would have been the norm when the early Bud- dhist texts were composed and transmitted, it would indeed seem that stratification of the early Buddh- ist literature is possible. What appear to be different strata according to the text critical method, may in fact have been formed through the vicissitudes of oral performance, perhaps because of a singer’s inclin- ation on any given day. According to Alexander Wynne, the views of Gombrich and of Norman seem to be based on intuit- ion rather than the internal evidence of the texts themselves. Cousins, too, does not cite much textual evidence to support his claim, though he does present some arguments. For example, he proposes that the material in the Sutta Piṭaka was formed by singers performing orally on “uposatha day or for the occasion of some sangha meeting,” or “when visiting the sick or for recitation after receiving food at the house of a layman.”78 Nevertheless, Wynne asks, what evidence is there in the early texts to support these different views?79 On the other hand, says Mark Allon, “there is, in fact, much scope for changes to occur within a ‘rightly memorized tradition’,”80 which he carefully demonstrates in his ground-breaking study, Style and Function: A study of the dominant stylistic features of the prose portions of Pāli canonical sutta texts and their mnemonic function (1997a).81 He gives us a detailed description of how the early Buddhist oral tradition works, as we shall see below. [4.2.2]

78 Cousins 1983:4 f. 79 Wynne 2004:98-100. 80 Allon discusses this in detail in the conclusion to his PhD thesis, “Some stylistic features of Pāli canonical sutta texts and their mnemonic function” (U of Cambridge, Sep 1994), a rev ed of which was published as Style and Function, Tokyo 1997. 81 Studia Philologica Buddhica 12. Tokyo: International Institute for Buddhist Studies of the International College for Advanced Buddhist Studies, 1997.

18 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

4.2.1.5 The 2nd theory of oral tradition explains that the composition of a fixed text is memorised and transmitted verbatim82 [4.2.5.7]. The standardised wording or passages outlined here function as an aid to composition within both of these methods: whenever a particular approach needed to be portrayed, the wording was already available. In other words, these formulas acted as prefabricated building- blocks.83 In addition to this, the use of a standardised and predictable diction would also have aided the learning by heart and recitation of a large body of fixed material; that is, within a tradition of the composition and transmission of fixed texts this feature would have a mnemonic function.84

4.2.2 The approach formula

4.2.2.1 Mark Allon has carefully investigated a number of the most prominent stylistic features of early Buddhist texts, and more specifically, of the prose of the Dīgha Nikāya, the first collection in the Sutta Piṭaka.85 One of the most dominant characteristics of the prose portions of Pāli suttas is the use of standardised phrases or passages to express or depict a given concept, action or event. These standardised phrases have been variously called “formulas,” “clichés,” “stock expressions,” “stock phrases” and “stereotyped phrases.” Allon prefers the term “formula.”86 In this section, we will study his survey of the approach formula often found in the suttas of the Dīgha Nikāya.87 The narrative portions of the suttas contain numerous passages which depict someone approaching another person, and the phrases used to depict these approaches are formulaic. Such material is extensive and diverse: we will only look at his discussion of those approach-formulas which are based on the unit or phrase, yena … ten’upasamkami, upasamkamitvā, “x approached y, having approached (he did such and such).”

4.2.2.2 The material in the Dīgha Nikāya can be divided into 2 groups. The formulas of the 1st group depict someone approaching the Buddha, a monk or another person, and a monk approaching the

82 See Gombrich 1990b. 83 Cf J Gonda 1959:41-43. 84 Cf G von Simson, 1965:142 f; 1977:479 f. But as mnemonic means “aiding memory,” it therefore not only includes “aiding the remembering of what is memorised,” but also “aiding the remembering of non-memorised elements,” for example, the course of events or the appropriate, or approximate, wording. We therefore find formulas and other elements of composition in an improvisatory setting being referred to as “mnemonic aids” (B A Ros- enberg, 1987:82 f) and “mnemonic elements” (J D Smith, 1989:40). It is used in this paper in the sense of “aiding the learning and recall of a memorised test” (A B Lord, 1987:67, makes the distinction between remembering and memorising; cf J Smith, 1989:36 f). The stylistic features discussed in this paper may have had other functions besides aiding composition. However, space does not permit a discussion of these here. See Allon 1997b:45 f. 85 The most important studies of the style of Buddhist or Brahminical literature are H Oldenberg, Zur Geschichte der altindischen Prosa, 1917: esp pp. 39-52; G von Simson, Zur Diktion einiger Lehrtexte des buddhistischen San- skritkanons, 1965; J Gonda, Stylistic repetitions in the Veda, 1959. On a smaller scale of study are S Weeratunge, “Some significant stylistic traits of Buddhist Sanskrit prose,” 1992, and P Kwella, “Some remarks on the style of some Buddhist Sanskrit texts,” 1978. For the Allon’s findings, 1997b:43-47 86 This diversity of terminology, says Allon, “in part reflects a general uncertainty as to what actually constitutes a ‘formula,’ ‘stock phrase,’ and so on, and whether, say, a ‘formula’ is different from a ‘stock phrase’ or ‘stock expression.’ A detailed discussion of the terminology and definitions is presented in his PhD theses. (1997b:56 n40) 87 Allon explains that “these studies were restricted to an analysis of prose, because, as stated, this is by far the most dominant medium used by the early Buddhists. It is also particularly interesting as most oral literature is verse. I chose to work with Pāli canonical sutta texts, not because they are the oldest, but because they represent the most complete and best preserved body of texts representative of the early phase of Buddhist literature.” (1997b: 56 n39) http://dharmafarer.org 19 SD 58.1 The Oral Tradition of the Early Buddhists

Buddha or another monk. The simplest formula within this group depicts the approach of the visitor, then his or her interaction, usually verbal, with the person approached. In the more complicated formulas, the visitor approaches, shows respect, adopts a particular posture (standing or sitting), then speaks with the person approached. A particular range of fixed units of meaning is employed within each division of this overall structure to construct distinct formulas. Which units of meaning are employed, and hence which formula type and specific formula is used, depends on the narrator, the classification of the person approaching and the person approached, their attitude towards each other, and the purpose of the visit.

4.2.2.3 So, for example, when a brahmin approaches a king, the following combination of units will be used:

(Then)88 the brahmin + approached the king. Having approached + he said this to the king … .89

In contrast, the formula used when 2 brahmin youths (Vāseṭṭha and Bhāra,dvāja) approaches the Buddha in order to question him will be:

(So) the brahmin youths + approached the Blessed One. Having approached the Blessed One, they + exchanged courtesies and + sat down at one side. Seated at one side, the brahmin + said this to the Blessed One … .90

This is characterised by respect being shown and the showing of this particular form of respect, by the brahmin assuming the same posture as the Buddha—sitting down when the Buddha is sitting, and standing when he is standing—and often with the brahmin speaking first.91 There are also certain forms of address associated with this interaction. Again, the following combination of units will be used to depict a monk approaching the Buddha when he has been summoned by him:

(Then) the monk + approached the Blessed One. Having approached, + having saluted the Blessed One, + he sat down at one side. To the monk + seated to one side, the Blessed One said this.92

88 Atha kho, “then” or “so” is added depending on the context. The + sign refers to where the name/s have been omitted. The translations here have been kept literal but standardized throughout for the sake of uniformity. 89 Eg, Mahā Govinda S (D 19): atha kho bho mahā,govindo brāhmaṇo yena reṇu rājā ten’upasakami, upasaṅka- mitvā reṇuṁ rājānaṁ etad avoca. (D 19,39/2:237), SD 63.4. 90 Eg, Te,vijja S (D 13): atha kho vāseṭṭha,bhāra.dvājā māṇavā yena bhagavā ten’upasaṅkamiṁsu, upasaṅkami- tvā bnhagavatā saddhiṁ sammodiṁsu, sammodanīyaṁ kathaṁ sārāṇīyaṁ vītisāretvā ekam antaṁ nisīdiṁsu ekam antaṁ nisinno kho vāseṭṭho māṇavo bhagavantaṁ etad avoca. (D 13,8/1:236), SD 1.8.. 91 However, this is not the rule, since the Buddha is reputed to be friendly by taking the initiative to “greet others (first)” (ehi,svāgata,vādī or -sāgata-), (a monk) who (habitually) says “come, you’re welcome" (to others), ie, a cult- ured monk; ehi sāgataṁ iti vadana,sīlo ti ~îti samāso, Sadd 743,18; amhākaṁ pana ayyā ... mihita,pubbaṅ,gamā ~ino (-svā-) abbhā,kuṭikā attāna,mukhā pubba,bhāsino, V 2:11,4 = 3:181,9 (~ino upāsakaṁ disvā "ehi svāgataṁ tavā’ti evaṁ-vādino, VA 622,34 f); samaṇo khalu bho gotamo ... ~ī (-sā-), D 1:116,9 (~îti deva,manussa,pabbajita,- gahaṭṭhesu taṁ taṁ attano santike āgataṁ “ehi sāgatan’ti evaṁ-vādîti attho, DA 287,1-2) = 132,20. The phrase occurs at D 4,6/1:116,9 (SD 30.5); D 5,7(18)/1:131 (SD 22.8); SD 45.16 (2.5.3(2)); of other monks: V 2:11, 3:181. 92 Eg Mahā,parinibbāna S (D 16): āyasmā ānando … yena bhagavā ten’upasaṅkami, upasaṅkamitvā bhagavan- taṁ abhivādetvā ekam antaṁ nisīdi, ekam antaṁ nisinnaṁ kho āyasmantaṁ ānandaṁ bhagavā etad avoca. (D 16,5.13/2:144), SD 9.

20 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

In contrast to the previous formula, this is characterised by the monk (Ānanda) showing this particular form of respect and by the Buddha speaking first. Again, there are certain forms of address found in such an approach.

4.2.2.4 The 2nd group of formulas depict the Buddha approaching someone and a monk approaching someone other than the Buddha or another monk. Here the situation is quite different. Not only do the structures of these formulas differ from those of the previous group, but these formulas also use a completely different range of fixed units of meaning. Although the factors which determine the use of particular fixed units of meaning, and hence the overall formula, are the same, their relative importance differs markedly from the 1st group. Here the purpose of the approach is the basic determinant of the formula used, with the wording of approaches of different purposes differing greatly from each other. We have, in effect, “purpose built” formulas. So, for example, the formula used to depict the Buddha visiting an ascetic (the wanderer Poṭṭhapāda) is:

Then the Blessed One approached the ascetic +. Then, the ascetic said this to the Blessed One: “Bhante, let the Blessed One come! Bhante, the Blessed One is welcome! It has been a long time, bhante, since the Blessed One has made this exception [gone out of his way], that is, in coming here. Bhante, let the Blessed One take a seat. Here is a seat that has been prepared.” The Blessed One sat down on the prepared seat, and the wanderer Poṭṭhapāda, taking a low seat, sat down at one side. The Blessed One said this to the ascetic seated to one side.93

4.2.2.5 However, the formula showing the Buddha attending a donor’s meal is thus:

Then, the Blessed One, having dressed in the morning and taking robe and bowl … went to the brahmin’s + residence. Having gone there, he sat down on the prepared seat. Then, the brahmin + with his own hands, served the community of monks headed by the Buddha, exquisite food, hard and soft, and satisfied them. When the Blessed One had finished his meal and taken his hand out of his bowl, the brahmin + sat on one of the low seats at one side. To the brahmin + who was sitting thus at one side, the Blessed One said this … .94

93 Eg Poṭṭhapāda S (D 9): atha kho bhagavā yena samaya-p,pavādako tindukā’cīro eka,sālako mallikāya ārāmo ten’upsaṅkami … atha kho bhagavā yena poṭṭhapādo parubbājako ten’upasaṅkami. atha kho poṭthapādo paribbā- jako bhagavantaṁ etad avoca: ‘etu kho bhante bhagavā, sāgataṁ bhante bhasgavato, cirassaṁ kho bhante bhag- avā imaṁ pariyāyaṁ akāsi yad idaṁ idh’āgamanāya, nisīdatu bhante bhagavā, idaṁ āsanaṁ paññattan’ti. Nisīdi bhagavā paññatte āsane. poṭṭhapādo pi kho paribbājako aññataraṁ nīcaṁ āsanaṁ paññattaṁ gahetvā ekam antaṁ nisīdi. ekam antam nissinnaṁ kho poṭṭhapādaṁ paribbājakaṁ bhagavā etad avoca. Note that the ascetic addresses the Buddha respectfully as bhante, “sir,” as the Buddhists would. (D 9,5/1:178 f), SD 7.14. 94 Eg Lohicca S (D 12): atha kho bhagavā pubba,samayaṁ nivāsetvā patta,cīvaraṁ ādāya saddhiṁ bhikkhu,saṅ- ghaṁ yena sālavatikā ten’pasaṅkami … atha kho bhagavā yena lohiccassa brāhmaṇassa nivesanaṁ ten’upasaṅka- mi, upasaṅkamitvā paññatte āsame nisīdi. atha kho lohicco brāhmaṇo buddha,pamukhaṁ bhikkhu,saṅghaṁ paṇīt- ena khādaniyena bhojaniyena sahatthā santappesi sampavāresi. atha kho lohicco bhagavantaṁ bhuttāviṁ onīta,- patta,paṇiṁ aññataraṁ nīcaṁ āsanaṁ gahetvā ekam antaṁ nisīdi. ekam antaṁ nisinnaṁ lohiccaṁ brāhmaṇaṁ bhagavā etad avoca. (D 12,7-9/1:226 f), SD 34.8. http://dharmafarer.org 21 SD 58.1 The Oral Tradition of the Early Buddhists

The formulas of this category are generally characterised by the following features:

(1) when the visit is a public one, the Buddha or the monk is depicted getting dressed and taking bowl and robe along; (2) the approach is to the place of the person who is approached rather than to the actual person; (3) the Buddha or monk does not show respect to the person visited; rather, some gesture of respect or subordination is shown by the person approached; (4) the Buddha or monk sits down on a prepared seat; and finally, (5) such approaches usually is reported by the sutta-narrator (in the 3rd person). Features 2, 3 and 4 tend to subordinate the person being approached to the Buddha or monk who is approaching.

4.2.2.6 The study of the material in the Dīgha Nikāya shows that the wording of passages which depict the common event of someone approaching another person has been standardised in this text to the extent that only a limited range of stock phrases or formulas is used.95 These formulas have set structures and are composed of a variety of possible fixed units of meaning. As mentioned, which units are employed, and hence which formula type and specific formula is used, depends on certain factors. Given some knowledge of these determining factors, the wording of a particular approach is, in the majority of cases, predictable.96 At the very least, this indicates that there is an overall homogeneity to the narrative portions of this collection of suttas.97 Whether there was a tendency to use a standardised diction from the beginning, or whether stand- ardisation was undertaken at the great councils (saṅgīti), or later by the reciters (bhāṇaka) tradition, or when the canon was written down, is yet to be determined.

4.2.3 Adaptability of the early oral tradition

4.2.3.1 Mark Allon, as we have noted, gives a comprehensive analysis of the nature of the structure and language of the early Buddhist texts in his study, Style and Function (1997a) [4.2.1.4]. In his paper, “The oral composition and transmission of early Buddhist texts” (1997b), he gives, as an example, a comparative study of the formulaic wording of the passages recounting Māra approaching the Buddha at the end of his life, as found in the suttas and the Vinaya.98 He starts by examining the related passage in the Dīgha Nikāya (D), the Udāna (U), the Saṁyutta Ni- kāya (S) and Aṅguttara Nikāya (A).99 The Dīgha Nikāya and the Udāna passages give the fullest formula with Māra approaching the Buddha, standing to one side, then speaking:

95 The situation is generally the same in other suttas, though the formulas used may differ [4.2.3]. 96 “The examples not mentioned in this paper of complicated and particularly detailed approaches, and of those which do not quite conform to the norm, show that these structures were not blindly imposed upon the material. The authors of this material were fully capable of breaking with the norm where necessary. Meaning was still the ultimate determinant of diction.” (Allon 1997b:57 n48) 97 Cf B J Manné, Debates and case histories in the Pāli canon. Unpubl PhD thesis, Rijksuniversiteit te Utretcht 1992:163. 98 We here give the main ideas of Allon’s analysis, adding details and further comments for a better understanding of the nature of the early Buddhist oral tradition. 99 Mahā,parinibbāna S (D 16,3.7/2:104), SD 9; Āyu,saṅkhār’ossajana S (U 6.1/51/63); Cetiya S (S 51.10/5:260); Bhūmi,cāla S (A 8.70/4:310); Nett 60.

22 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

Then, not long after the venerable Ānanda had left, Māra the bad approached the Blessed One. Having approached, he stood to one side. Standing to one side, Māra the bad said this to the Blessed One … .100 (D 2:104, translation standardized throughout), SD 9

4.2.3.2 The Saṁyutta Nikāya has a briefer formula with no mention of Māra standing to one side:

Then, not long after the venerable Ānanda had left, Māra the bad approached the Blessed One. Having approached, he said this.101 The Aṅguttara Nikāya passage is, however, so brief that it does not even mention the approach, merely reading:

Then, not long after Ananda had left, Māra the bad said this to the Blessed One ... .102

The D and U texts are the most detailed, highlighting the narrative aspect of the occasion: the lesson is brought out by the story as a whole. This is a natural tendency in the D suttas, since the purpose of the Dīgha is to serve as a kind of broad prospectus to impress and attract the non-Buddhist with what Buddh- ism has to offer.103 The A texts are mostly for the instruction of the laity and neophyte: the familiar narrative of Māra is severely stripped down to instead highlight the sutta’s teaching.

4.2.3.3 The D formula or stock passage depicting the Buddha or a monk approaching an ascetic is characterised by the ascetic showing respect to the Buddha or the monk, and not the other way around. A well known case of this is found in the Udumbārika Sīha,nāda Sutta (D 25),104 where the wanderer, Ni- grodha, leader of 3,000 wanderers, shows deference to the layman Sandhāna and to the Buddha by cour- teously offering a prepared seat, and so on. Such gestures tend to subordinate the ascetic to the Buddha or the monk or, in this case, a Buddhist layman.105 [4.2.2] Further research by Allon shows that this respect formula is specific to the Dīgha Nikāya and the Majjhima Nikāya (M). In contrast, the S, A and V formulas are simpler, depicting the Buddha or the monk greeting the ascetic, sitting down on the prepared seat, then speaking to him.106 When depicting such en- counters with ascetics, the compilers of S, A and V do not portray the Buddha or monk being honoured in such an elaborate manner.

4.2.3.4 There are differences between the various sutta texts in their wording of the “going to an invited meal” approach-formula mentioned earlier. Although the same basic formula is used throughout the suttas, the syntax of the fixed units of meaning and the presence of particular units differs from text to text. For example, in D and U, the Buddha approaches the donor’s house “together with a community of monks” and sits down. However, in M, A, Sn (Sutta Nipāta) and V, we have the Buddha approaching the

100 Atha kho māro pāpimā acīra,pakkante āyasmante ānande yena bhagavā ten’upasaṅkami, upasaṅkamitvā ekam antaṁ aṭṭhāsi. Ekam antaṁ ṭhito kho māro pāpimā bhagavantaṁ etad avoca (D 16,3.7/2:104), SD 9. 101 [56] Atha kho māro pāpimā acīra,pakkante āyasmante ānande yena bhagavā ten’upasaṅkami, upasaṅkami- tvā etad avoca (S 5:260). 102 [57] Atha kho māro pāpimā acīra,pakkante āyasmanti ānande bhagavantaṁ etad avoca (A 4:310). 103 See SD 30.8 (4.1.2). 104 The wanderer Nigrodha, leader of 3000 wanderers, shows deference to the layman Sandhāna and the Buddha (D 25/3:36-57), SD 1.4. 105 Allon 1997b:43-45. 106 [58] Eg S 2:32 f. Atha kho āyasmā sāriputto yena añña,titthiyānaṁ paribbājikānaṁ ārāmo ten’upasaṅkami, upasaṅkamitvā tehi añña,titthiyehi paribbājakehi saddhiṁ sammodi sammodanīyaṁ kathaṁ sārāṇīyaṁ vītisāretvā ekam antaṁ nisīdi. Ekam antaṁ nisinnaṁ kho āyasmantaṁ sāriputtaṁ te añña,titthiyā paribbājakā etad avocuṁ. http://dharmafarer.org 23 SD 58.1 The Oral Tradition of the Early Buddhists donor’s house, then sitting down “together with the community of monks”—this is associated with the verb for sitting down rather than with the verb for approaching. Clearly, this shows the Buddha and his monks as the honoured guests.107 Also, according to the D, M, U and Sn formulas, the donor, “taking a lower seat, sat down to one side” after the meal, while the S, A and V formulas do not include this “taking of a lower seat” phrase.108 Clearly, S and A suttas highlight the Dharma aspect; hence, the social formalities are downplayed. V is generally late, and tends to echo passages from the older sections of the Pali canon.

4.2.3.5 The situation seems to be quite complex with a text like the Udāna, for example, following the wording (diction) of D with regard to certain formulas. It is possible that such differences may have resulted from the “reciter” (bhanaka) tradition or the period of specialisation. Alternatively, differences in wording may have resulted from the way in which each text was used by the Buddhist community. In other words, it is possible that different texts were intended for different audiences and had different functions, and that their wording was modified accordingly. Or again, in some cases, these differences may have arisen from the different manuscript traditions (that is, the Pali text of the suttas). In other words, it depends on whether the sutta manuscript was from Sri Lanka, Myanmar, Thailand, Cambodia (Khmer) or Laos; and, further, within each country, there are various ancient manuscripts, often with variant readings. Translators and scholars depend on as many good manuscripts as they can access. Then again, the more manuscripts there are, the more collations and comparisons have to be diligently made. Anyway, for effective sutta work, a handful of the best ancient manuscripts would often well serve the purposes of translation and sutta editing. Nevertheless, further research clearly needs to be done to properly identify and understand such differences in the readings and textual formulas.

4.2.4 Sound similarities: The formal sutta opening109

4.2.4.1 All the suttas in the Dīgha Nikāya and the Majjhima Nikāya, at the start of sutta groupings in the Saṁyutta Nikāya and the Aṅguttara Nikāya, and many of the suttas in the Khuddaka Nikāya, formally open with “Thus have I heard. At one time … ,” (evaṁ me sutam. ekaṁ samayaṁ), followed by saying where the Buddha is at that time (that is, at the start of the sutta account or during its account). The opening “I have heard” (me sutaṁ, literally, “heard by me”) clearly highlights the oral nature of what is to follow.110

107 [59] Eg U 82, 89 read atha kho bhagavā pubbaṇha,samayaṁ nivāsetvā patta,cīvaraṁ ādāya saddhiṁ bhikkhu,saṅghena yena [house of host] ten’upasaṅkami, upasaṅkamitvā paññatte āsane nisīdi, while M 2:146, Sn p111, A 4:187 and V 1:217 f read atha kho bhagavā pubbaṇha,samayaṁ nivāsetvā patta,cīvaraṁ ādāya yena [house of host] ten’upasaṅkami, upasaṅkamitvā paññatte āsane nisīdi saddhiṁ bhikkhu,saṅghena. 108 Eg D 2:97 reads atha amba,pālī gaṇikā buddha,pamukhaṁ bhikkhu,saṅghaṁ pāṇitena khādaniyena bhojaniyena sahatthā santappesi sampavāresi. Atha kho amba,pālī gaṇikā bhagavantaṁ bhuttāviṁ onīta,patta,pāṇiṁ aññataraṁ nīcaṁ āsanaṁ gahetvā ekam antaṁ nisīdi, while its parallel at V 1:233 reads atha kho amba,pālī gaṇi- kā buddha,pamukhaṁ bhikkhu,saṅghaṁ khādaniyena bhojaniyena sahatthā santappetvā sampavāretvā bhaga- vantaṁ bhuttāviṁ onīta,patta,paṇiṁ ekam antaṁ nisīdi. 109 This section is based on Analayo, “Oral dimensions of Pāli discourses,” 2007, ie, with some modifications and elaborations. 110 Levering also notes another function of this introductory formula, in that “teachings were authenticated by the fact that one could demonstrate that ... they had been heard by a specific hearer, that he had heard the Bud- dha teach them at a particular time and place.” (“Scripture and its reception,” 1989:61). Cf the opening of the suttas of the Iti,vuttaka, said to be memorized by the lay follower, Khujj’uttarā: vuttam h’etaṁ bhagavatā vuttam

24 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

According to the traditional account, these words were spoken by elder Ānanda, the Buddha’s personal attendant during the last 25 years of the ministry. They stand for his oral reception and subsequent transmission of the teachings he had heard (MA 1:7). Not only the content, but also the form of this formulaic opening testifies to oral transmission.

4.2.4.2 In regard to this formal sutta opening, however, Brough follows the Tibetan version: ‘di skad bdag gis thos pa dus gcig na | bcom ldan ‘das, in taking ekaṁ samayaṁ as qualifying evaṁ mayā srutaṁ, that is, “at one time I heard: the Blessed One was staying at ... .”111 In fact, the Tibetan simply back-trans- lates to Sanskrit as evaṁ mayā śrutaṁm ekasmin samaye bhagavān | … viharati sma, and the Chinese is 如是我聞一時佛在舍衛 “Thus have I heard. Once the Blessed One was dwelling (in) … “ (Vajracchedika Prajñāpāramitā Sūtra).112 Hence, his view has mostly been rejected by other scholars. For Hinüber, in as much as Pali sources are concerned, no arguments can be found in support of assuming that ekaṁ samayaṁ qualifies evaṁ me sutaṁ.113 Tola, too, disagrees and thinks that to use the qualification “at one time” in regard to the Buddha’s whereabouts seems more meaningful than to use the same qualification to indicate that the oral transmission of the sutta took place “at one time.”114 Moreover, says Analayo, the phrase tena samayena, “at that time,” that regularly introduces the next sentence in the formal beginning of a sutta clearly refers to the time when the events mentioned in the sutta occurred, in view of which it would be more natural for the preceding ekaṁ samayaṁ to refer to the same.115

4.2.4.3 Even among these first few words, found at the beginning of each sutta, sound similarity and metrical identity are evident, and these are used in all the suttas and texts to facilitate recitation and memorization. Such sound similarities involve alliteration, repetition of an initial sound, assonance, repetition of a sound found in the middle of a word, and homoioteleuton, repetition of the final sound. As shown in Figure 4.2.4, the two parts of the standard sutta opening, evaṁ me sutaṁ and ekaṁ samayaṁ, each consists of 5 syllables. The first word in either part is closely similar, evaṁ and ekaṁ, differing only in their second consonant. The words evaṁ, sutaṁ, ekaṁ and samayaṁ share the -aṁ ending,116 while the words sutaṁ and samayaṁ share the same initial consonant.117 Thus, even though these few words are merely a prose introduction to a sutta, a closer examination reveals sound similarities and rhyming that occur with considerable frequency in other prose sections of the early discourses, especially in listings of similar words or in formulaic expressions. The alliteration in the 1st word of the 2 sentences refers to the same -aṁ ending in either of them, giving them a final rhyming. The assonance shows the same -w- sound of -v- = -u- as the middle sound. arahatā’ti me sutaṁ, “This was indeed spoken by the Blessed One, heard by me spoken by the Arhat.” See SD 57.26 (1.2.2.3); SD 16.14 (1). 111 “Thus have I heard … ,” 1950:416. 112 https://www.ayurveda-institute.org/ayurvedic-medicine-online- course/doku.php?id=vajra_sutra_verses_comparison. 113 Studien zur Kasussyntax des Pali, 1968:85 f. 114 “Ekaṁ samayam,” 1995:54. 115 Analayo 2007:25 n1. For a criticism of Brough’s arguments, see also Galloway, “Thus have I heard: At one time … ,” 1991 and Klaus, “Zu der formelhaften Einleitung der buddhistischen Sutras,” 2007. Samtani notes that Jain sutras have a similar opening: suyam me (“The opening of the Buddhist Sūtras,” 1964:49). Further see Bon- gard-Levin, “The Nagaropamasūtra,” 1996:90 n1. 116 Analayo thinks that the choice of the acc ekaṁ samayaṁ instead of the loc ekasmiṁ samaye (as in the Skt) might even be related to the sound similarity this creates with the preceding evaṁ me sutaṁ. Cf the gloss at MA 1:10 as tasmiṁ samaye and Wijesekera 1993:56. 117 Allon 1997a:195, 242. http://dharmafarer.org 25 SD 58.1 The Oral Tradition of the Early Buddhists

The assonance is further balanced with 5 syllables in either words: evaṁ me sutam and ekaṁ samayaṁ. In Pali both the semi-vowel v and the vowel u have the same sound value. The homoioteleuton gives the same initial s-. Hence, we have here a case of full rhyming, initial, medial and final!

evaṁ me sutaṁ or ekaṁ samayaṁ evam me sutaṁ alliteration: same -aṁ ending evaṁ ekaṁ assonance: same middle sound evaṁ … sutaṁ ekaṁ samayaṁ homoioteleuton: same initial s- sutaṁ samayaṁ

Figure 4.2.4: Sound features in the two sentences of the formal sutta opening

4.2.5 Waxing syllable principle

4.2.5.1 Scholars have noticed another common feature in the prose portions of the suttas: this is its tendency to proliferate similar words and phrases (“word elements and meaning units”) to form sequences or “strings.”118 We frequently see sequences of 2, 3 or more adjectives or adjectival phrases qualifying the same noun, a number of nouns, all the subjects of the same sentence, or the objects of the same verb. We see strings of adverbs modifying the same verb, or a sequence of related verbs in the same sentence, and so on. Wherever such sequences of parallel words and phrases occur, they are arranged according to what philologists call the rule of waxing components or the waxing syllable principle.119 This is a sequence of related words and phrases of an unequal number of syllables, where (expanding on Paṇini’s terminology) those of fewer syllables must precede those of more syllables, giving a euphonic flow of sounds.120

4.2.5.2 The waxing syllable principle is frequently applied to strings of synonyms. These strings of synonyms serve to safeguard against textual and memory loss. They prevent a physical loss of text and personal loss of teaching. A whole string of similar words stands a much greater chance of being remembered than a single word; a string of synonyms impresses itself on the audience prone to distraction: just catching a single synonym is likely to help the audience keep track of what is being taught.121 A closer look at such strings of words brings to light that its members tend to occur in a metrical sequence that follows the principle of waxing syllables. As we have noted, by this principle, in a series of words, those with fewer syllables are followed by words with an equal or greater number of syllables. This principle is also applied to listings and enumerations whose members do not share the same mean-

118 This section on the waxing principle is based on Allon 1997b:47-50 and Analayo 2008:5-9. 119 Das Gesetz der wachsenden Glieder. See H Ehelhof, Ein Wortfolgeprinzip im Assyrisch-Babylonishen, Leipzig, 1916; Leumann-Hofmann 804 f. See CPD (1:35*): wax comp; Analayo, in Ency Bsm 8: waxing syllables: https://www.buddhismuskunde.uni-hamburg.de/pdf/5-personen/analayo/encyclopedia-entries/waxing.pdf. 120 Pāṇini (fl 400-350 BCE) was an ancient Sanskrit philologist, grammarian and scholar in what is today Pakistan. His classic is The Ashṭādhyāyī, 1962; Astadhyayi of Panini: In Roman Transliteration (Texas Linguistics Series), 1987; The Aṣṭādhyāyī of Pāṇini with translation and explanatory notes, 1991. See W Caland, “A rhythmic law in language,” 1931:59; cf J Gonda, Stylistic Repetitions in the Veda, 1959:61. 121 Oldenberg comments that the use of such strings of synonyms gives the impression of a certain childlike in- sistence that ensures that all aspects of a particular matter find expression. (1917:42)

26 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists ing. A few selected examples in Figure 4.2.5 show how a particular theme is expressed by a string of words with ascending syllable count.122

theme Pāli terms syllable count old jiṇṇo vuddho mahallako addha,gato vayo,anuppatto 2+2+4+4+6 growth vuddhiṁ virūḷhiṁ vepullaṁ 2+3+3 fear bhīto saṁviggo loma.haṭṭha,jāto 2+3+6 to (mis)meditate jhāyanti pajjhāyanti nijjhāyanti apajjhāyanti 3+4+4+5 able to attain nikāma,lābhī akiccha,lābhī akasira,lābhī 5+5+6 poor daliddo assako anāḷhiyo 3+3+4 wealthy aḍḍho mah’addhano mahā,bhogo 2+4+4

Fig 4.2.5: The waxing syllable principle (Analayo)

The crescendo effect that results from the application of this principle is a typical stylistic feature of the early suttas, further enhanced when word sequences follow the waxing syllable principle also share sound similarities. If a sequence of words becomes relatively long, this principle is not applied to the sequence as a whole, but to subunits within the sequence. Such subunits can share a similar nuance of meaning or belong to the same category, and the division into subunits may have the function to set a rhythm that allows the reciter to take a breath in between sutta recitation. An example of this is the description of various types of “animal talks,” which will be examined next. [4.2.5.4]

4.2.5.3 We see a common example in, for example, the Udumbarikā Sīha,nāda Sutta (D 25), when the wanderer Nigrodha, who arrogantly challenges the Buddha, but, in the end, confesses that he is “just as one foolish, as one confused, as one unskilful”: yathā,bālaṁ yathā,mūḷhaṁ yathā,akusalaṁ.123 The phrase consists of a string of 3 adverbial phrases. The first has 4 syllables, the second 4 syllables and the third 6 syllables; forming the pattern 4+4+6. Its 1st occurrence in the Sutta, in the instrumental mode (rare)—yathā,bālena yathā,mūḷhena yathā,- akusalena—forming the pattern 5+5+7, also follows the same principle of placing components with more syllables later. This arrangment of components according to an increasing syllable length tends to produce a crescendo effect, facilitating a smoother enunciation for a euphonic recitation in an oral tradition.

4.2.5.4 An exception to this principle is where a sequence, and especially a long sequence, can or must be divided into groups on the basis of associations in meaning, or grammatical or morphological

122 Examples from Majjhima Nikāya vol 1, esp from M 12/1:82,26, 16/1:101,7, 35/1:231,37, 50/1:334,23, 53/- 1:354,36, 66/1:450,34 + 1:451,36, foll Hinüber 1994b:16-30 and CPD 35*. Hinüber 1994b:33 points to similar formulas in Jain scriptures, such as AMg naṭṭā, gīa, vāiya, corresponding to Pali nacca, gīta, vādita found eg at M 27/- 1:180,6; cf Allon 1997a:266. 123 D 25,22/3:55,5+9 (SD 1.4); also Sāmañña,phala S (D 1:85x2); Bhaddāli S (M 1:438x2, 439x2, 440x2); Dhātu Vi- bhaṅga S (M 3:246x2); Ujjhāna Saññi S (S 1:24; vl -ā ending); Susima S (S 2:127x2); Ovāda S (S 2:205x2; vl -e ending); Saṅkavā S (A 1:238x2); Bhikkhuṇī S (A 2:146x2); Sīha,nāda S (A 4:377x2); V 2:126x2, 192x2, 4:18x2. Ins, yathā,bālena yathā,mūḷhena yathā,akusalena (D 25,21/3:54,1). Loc, yathā,bāle yathā,muḷhe yathā,akusale (V 1:315x2). http://dharmafarer.org 27 SD 58.1 The Oral Tradition of the Early Buddhists form, in which case the Waxing Syllable Principle only works within each group of words, beginning again with each group. For example, the stock list of the “animal talk” (tiracchāna,kathā) engaged in by ascetics, which also occurs in the Udumbarikā Sīha,nāda Sutta, consists of a long list of conversation topics. The following sequence of phrases from the Sutta can be divided into groups on the basis of associations in meaning, with the numbers showing the numerical sequences of each group or phrases of the “animal talk” pericope, thus: raja,kathaṁ cora,kathaṁ mahâmatta,kathaṁ (“talk of kings, thieves, ministers”) 4+4+6 senā,kathaṁ bhaya,kathaṁ yuddha,kathaṁ (“talk of armies, fear, battle”) 4+4+4 anna,kathaṁ pāna,kathaṁ vattha,kathaṁ sayana,kathaṁ (“talk of food, drink, clothing, beds”) 4+4+4+5 mālā,kathaṁ gandha,kathaṁ ñāṭī,katham yāna,katham (“talk of garlands, scents, relatives, vehicles”) 4+4+4+4 gāma,-kathaṁ nigama,-kathaṁ nagara,kathaṁ janapada,kathaṁ (“talk of villages, towns, cities and districts”) 4+5+5+6 and so on (D 3:36 f).124 This principle of component sequencing is not only apparent in the more obvious sequences of adjectives and adjectival phrases, nouns and noun phrases, adverbs and verbs, but it also seems to be oper- ational in the ordering of parallel units of meaning which occur in different, but closely associated, claus- es, sentences and paragraphs, and in the ordering of sequences of parallel sentences or semi-independent phrases, as well as a number of other structures, such as those involving the conjunction, saddhiṁ, “with.”

4.2.5.5 The waxing syllable principle not only applies to long phrases, but also to short ones, such as the shortest phrase, the dvanda (twin) compounds. Take for example the pacittiya rule prohibiting a monk from teaching one not ordained more than “6 or 5,” cha,pañca (V 4:21,37): the Pali sequence of numbers clearly follow the principle of waxing syllables against the natural ascending order of the numbers 5 and 6 (as would be the case in English). Another example is the well-known expression dhamma,vinaya. If we follow the sequence of the 3 trainings, those in moral virtue (sīla,sikkhā), in concentration (samādhi,sikkhā) and in wisdom (paññā), then, we have the traditional arrangement of the Tipiṭaka as Vinaya, Sutta and Abhidhamma. In this well known dvandva, the Vinaya is placed second, following the principle of waxing syllables: two-syllabled dhamma precedes 3-syllabled vinaya; hence dhamma,vinaya. In fact, the application of the waxing syllable principle to dvanda compounds is a rule recognized by Pāṇini himself.125

4.2.5.6 Most of the longer stock phrases, like those discussed above, follow the waxing syllable principle [4.2.5.1]. However, there are exceptions with problematic patterns. Such situations can be corrected by vowel insert or anaptyxis (Skt svarabhakti),126 or a word thought to be a later insertion is omitted, or when only the immediately parallel units are compared. Further, conceptual considerations or the desire to produce a particular word-play may occasionally override the waxing syllable principle.

124 See Allon 1997b 48 f; Analayo 2007:7 f. 125 Caland 1931:59-68 quotes Pāṇini 2.2.34 and gives a list of examples, such as strī,pumāṁsu, “female and male,” and kanyā,kumārau, “maiden and youth,” showing how the waxing syllable principle applies to dvanda cpds giving a sequence that seems to go against their natural order. 126 In comparative philology and phonology, a svarabhakti is where a vowel is inserted between two consonants; a kind of epenthesis. Sometimes known as a “parasitic vowel,” such as in the pronunciation of film as |ˈfilǝm| in some English dialects.

28 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

It is particularly common in these sequences for the components, and especially the initial components of the sequence, to share sound and metrical similarities.127 In the above example of yathā,bālaṁ yathā,mūḷhaṁ yathā,akusalaṁ [4.2.5.2], it is seen that, apart from the obvious sound similarities due to yathā- being the first member of each compound, the endings of all three are virtually identical: they all end in -aṁ. And in the case of yathā,bālena yathā,mūḷhena yathā,akusalena, their endings rhyme with -ena, as -lena, -ḷhena, lena. In fact, the first two phrases of this sequence differ only in their core syllables: -bā- and -mū-, which are both labial consonants with long vowels. The first two phrases therefore share the same metrical measure (2 counts) and are virtually the same word. The -ū- of the second member (-mūḷhena) is also rhymed in -akusalena of the third. Hence, there is a tendency in these texts to proliferate similar word elements and phrases, that is, to expand the wording and also linking them by sound and metre. In short, we have the features of poetic prose.

4.2.5.7 This phenomenon of ordering similar word elements according to their syllable lengths has been known for some time, but an analysis of the nature and extent of its application within Pali texts has not been fully undertaken before until Allon’s work (1997b).128 The proliferation of similar word elements and phrases, and the ordering of the member elements of such sequences according to the Waxing Syllable Principle, which thus produces an overall crescendo and rhyme effect, tends to give a rhythm and homogeneity to this material. This rhythm and homogeneity is greatly enhanced when, as is frequently the case, the member elements also share rhyming sound and the same metre. This rhyming patterns in suttas, on account of sound and metre, especially in long prose texts, are extremely important in their recitation. As fixed texts, this use of formulas we have discussed, would act as an organizing principle, an aid in composing of such sutta texts. A combination of these stylistic features—sound and metre—in these sutta phrases (following the waxing syllable principle) function as a mnemonic aid: it is surely easier to remember a sequence of words arranged in this way according to syllable length. Similarly, it is easier to remember two different words when they share similar or rhyming sounds and same metre. The presence of such rhythm surely facilitates the memorisation and recitation of such sutta texts. Such choice of words and their arrangement were, of course, influenced by the fact that these texts were composed and transmitted orally.

4.2.6 Repetition (Mark Allon)

4.2.6.1 This section is based on Mark Allon’s detailed study of repetition in the Udumbarikā Sīha,- nāda Sutta (D 25), as summarized in his 1997b article.129 By repetition here is meant the recurrence of sentences, passages or whole sections of suttas, and the repetition of set structures in the suttas. In order to establish the degree to which Udumbarikā Sīha,nāda Sutta is repetitive, Allon had the sutta text of the PTS edition scanned into a word processor and all abridged passages (peyyāla) were laid out in full.

127 What mattered in an oral tradition are repetition and sound similarity. But specific kinds of sound similarities, found in English, such as alliteration, assonance, homoioteleuton, etc, are not important here, as noted by J Gonda (1959:376 f); but see (4.2.4.3). 128 See W Caland 1931; J Gonda 1959 eg 60-64, 125 f: G von Simson, 1965 eg §§2.3, 2.7, 8.4-6; H Smith, Epilego- mena to CPD 35* wax comp; O von Hinüber 1990b ch VII; 1993:104-113; 1994b, a very detailed study. 129 Mark Allon, Style and Function, 1997a; summarized in “The oral composition and transmission of early Buddh- ist texts,” 1997b (from which this section is taken). http://dharmafarer.org 29 SD 58.1 The Oral Tradition of the Early Buddhists

It was then possible to establish a word count for the complete Sutta and for those sections which were being repeated, and hence to calculate what percentage of the text was repetitive.130 In order to quantify repetition, the level at which the repetition occurs within the text and the type of repetition involved must be established. Repetition can occur at a number of levels. A passage is repetitive at a primary level when it does not form part of a passage which is itself repeated within the text. If it does, then it is repetition at a secondary level. Sometimes repetition at a tertiary level is discernible.

4.2.6.2 In Allon’s study, 5 categories or types of quantifiable repetition were established. These categories and the percentage (rounded) of the Sutta they each embody are as follows:

VR Verbatim Repetition 1,761 words 30% RMM Repetition with Minor Modifications 2,028 words 35% RIM Repetition with Important Modifications 222 words 4% RS-1 Repetition of Structure Type-1 937 words 16% RS-2 Repetition of Structure Type-2 149 words 3% Total 5,097 words

4.2.6.3 In Verbatim Repetition a passage is repeated word for word with no modifications needing to be made by the one who recites or performs this material. For example, the stock description of the animal-talk engaged in by ascetics mentioned earlier is repeated verbatim 4 times in this Sutta, representing about 5% (4.5%) of the text. Or a long passage which describes 3 stages of what the Buddha considers to be true ascetic practice, and which represents about 6% of the text, is repeated verbatim 3 times, making up about 17% of the Sutta. In total 30% of the Udumbarikā Sīha,nāda Sutta involves Verbatim Repetition at a primary level. Passages which are repeated with alteration to only a small proportion of their wording were classified as Repetition with Minor Modifications. For example, it is not uncommon to form the opposite of a passage expressing a positive or negative state by merely repeating that passage and adding or omitting certain prefixes or particle.131 Just under 35% (34.5%) of this Sutta involves this kind of repetition on a primary level. The third category, Repetition with Important Modifications, involves repetition of a passage, but with important changes to the wording, whether in syntax, grammatical number, tense or person, or enlargement or contraction of the wording. Material of this category represents about 4% (3.8%) of the Sutta studied.

4.2.6.4 Two types of repetition of structure were also established. In Repetition of Structure Type-1, a structure is repeated along with virtually all of its wording, but with key elements replaced to produce differences in meaning. So, for example, in the common passage which depicts the practice of the 4 divine abodes (brahma,vihara), the same structure and wording is repeated 4 times, first for lovingkindness

130 The complete test is 34% larger than the abbreviated PTS version. 131 For example, the passage puna ca paraṇ nigrodha tapassī tapaṁ samādiyati, bhojanesu vodāsam āpaj jati “idaṁ me khamati, idaṁ me na-k,khamatîti. So yaṁ hi kho’ssa na-k,khamati taṁ sāpekho pajahati, yaṁ pan’assa khamati taṁ gathito mucchito ajjhāpanno anādīnava,dassāvī anissataṇa,pañño paribhuñjati (D 3:43) is later repeated as puna ca paraṁ nigrodha tapassī tapaṁ samādiyati bhojanesu na vodāsaṁ āpajjati “idaṁ me khamati, idaṁ me na-k,khamatîti. so yaṁ hi kho’ssa na-k,khamati taṁ anapekho pajahati, yaṁ pan’assa khamati taṁ agathito amucchito anajjhāoanno ādīnava,dassāvī nissaraṇa,pañño paribhuñjati D 3:46).

30 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

(mettā), then for compassion (karuṇā), then joy (muditā), and finally, for equanimity (upekkhā) (eg D 3:49 f). In this way each repetition differs by only one word. In many passages of this class, the elements which differ in each repetition share morphological, sound or metrical similarities, or similarities in structure (or some combination of these), thereby minimising the impact of the changes being made. For example, in this Sutta, after defeating the ascetic Nigrodha in debate, the Buddha criticises him for not having the following thought:

Awakened (buddho), the Blessed One teaches the Dharma for awakening (bodhāya); tamed (danto), the Blessed One teaches the Dharma for taming (damathāya); calmed (santo), the Blessed One teaches the Dharma for calm (samathāya); crossed over (tiṇṇo), the Blessed One teaches the Dharma for crossing over (taraṇāya); extinguished (parinibbuto), the Blessed One teaches the Dharma for extinguishing (pannibbānāya).132

The initial element of each parallel sentence (buddho, danto, santo, tiṇṇo, parinibbuto) is a past participle. The 1st four have the same number of syllables and equal metrical patterns, and sound similarities are evident at least in danto and santo. In the 2nd group of elements which differ in each repetition, dam- athāya and samathāya are morphologically parallel, share the same number of syllables and have the same metrical pattern. They differ, in fact, only in their initial letter. Similarly, the last two elements in this group, taraṇāya and parinibbāṇāya, are morphologically similar. In this way, the elements which differ in meaning within each repeated structure appear similar in outward form. The effort involved in making the required modifications is thereby minimised for the reciter. A total of 16% of this Sutta is composed of material of this RS-1category.

4.2.6.5 In those passages which were classified as Repetition of Structure Type-2, a basic structure is repeated, but with far less repetition of the wording, or in some cases, with modification to the structure of the wording which is replaced. Material of this category represents nearly 3% (2.5%) of this Sutta. In total almost 87% (86.8%)133 of the Udumbarikā Sīha,nāda Sutta involves quantifiable repetition of one kind or another at a primary level. This is surely a significantly high proportion of the text. It must also be noted that the verbatim end of the scale is particularly well represented.

4.2.6.6 Repetition is undoubtedly a mnemonic device. This is based on the simple observation that the more frequently a passage, phrase or word is repeated, the more likely it is to be remembered. Or as a verse in the Dhammapada states: “Non-recitation is the rust of incantation” (asajjhāya,mālā mantā, Dh 241). The repetition encountered in Buddhist texts has frequently been taken to have a mnemonic function, but few have elaborated on their statements or investigated repetition in any systematic manner.134 Allon has so far argued that the first 2 stylistic features discussed above could have functioned as aids to composition both within a tradition of composing material during the performance in an improvisatory manner and in a tradition of composing fixed texts which were to be transmitted verbatim. In

132 [69] “tassa te nigrodha viññussa sato mahallakassa na etad ahosi: ‘buddho so bhagavā bodhāya dhammaṁ deseti, danto so bhagavā damathāya dhammaṁ deseti, santo so bhagavā samathāya dhammaṁ deseti, tiṇṇo so bhagavā taraṇāya dhammaṁ deseti, parinibbuto so bhagavā parinibbānāya dhammaṁ deseti’”(D 3:54 f). 133 [70] The complete Sutta has a word count of 5,871. The word count for the passages which have been classed as VR is 1,761. The word county for the other 4 categories are: RMM 2,028; RIM 222; RS-1 937; RS-2 149. The 5 categories have a total word count of 5,097, which represents 86.8% of the text. 134 [72] G von Simson (1965 esp 142 ff) is the exception. Cf T W Rhys Davids, Vinaya Texts vol 1, 1881:xxii f. R Gombrich 1990b:24; W B Bollée, 1970:172; L S Cousins 1983:9; Simson 1977:480; A Syrkin 1983:160; J Gonda 1959: 78, 351. http://dharmafarer.org 31 SD 58.1 The Oral Tradition of the Early Buddhists addition to this, it was proposed that these features would also have had a mnemonic function within the latter tradition. In contrast to this, it is difficult to see the gross forms of repetition just discussed—the repetition of whole passages, with or without modification, and the repetition of structures with the replacement of various proportions of their wording—and the scale on which this is pursued, that is the proportion of the text involved, as anything other than proof, or at least as a very strong indication, that these texts were designed to be memorized and transmitted verbatim.135

4.2.6.7 In contrast, material such as the contemporary Yugoslav epics studied by Lord (1960) or the contemporary Indian epics studied by Smith (1991), Beck (1982) and Roghair (1982), which is composed “during the performance,” although exhibiting many forms of repetition, does not exhibit the form of gross repetition encountered in Pali sutta texts. The 5 categories of repetition established in the study upon which the above discussion of repetition in the Udumbarikā Sīha,nāda Sutta is based can be graded according to the degree to which they each facilitate the learning and retention of this material. Verbatim Repetition obviously represents the great- est aid to memory. The greater the percentage of a text that is verbatim repetitive the easier it is to learn and remember. At a primary level, 30% of the Udumbarikā Sīha,nāda Sutta consists of repeated passages of this classification. The remaining four types of repetition each encompass a range of differences. In terms of the modifications to be made by the reciter, and hence the effort involved in making such changes, the Repe- tition with Minor Modifications and Repetition of Structure Type-1 categories on the one hand, and the Repetition with Important Modifications and Repetition of Structure Type-2 categories on the other, are seen to be parallel and to encompass a similar range of differences.

4.2.6.8 As mentioned, almost 35% of the Sutta studied involves Repetition with Minor Modifica- tions at a primary level. Another 16% involves Repetition of Structure Type-1. Together these two categories, which are similar in terms of their mnemonic significance, represent about 51% (50.5%) of this Sutta. The study also showed that almost 4% of this Sutta involves Repetition with Important Modifica- tions at a primary level and that approximately 3% involves Repetition of Structure Type-2. Together these two parallel categories represent about 6% (6.3%) of this Sutta. As 87% of the Udumbarikā Sīha,nāda Sutta involves some form of quantifiable repetition on a primary level, 13% of this text is therefore only encountered once. Much of this consists of the opening and closing sections of the Sutta. Although not occurring again, the passages and elements which make up this 13% commonly involve non-quantifiable forms of repetition (as do those which are repeated again) and may be found elsewhere in the Nikāyas.136 Many of the passages which are repetitive at a primary level in the text are themselves composed of or incorporate quantifiable repetitive elements, which is repetition at a secondary level. For example, a passage which is repeated verbatim may itself be composed of a passage repeated verbatim twice. This secondary passage therefore occurs four times in the text. This secondary repetition would further in- crease the familiarity of the material being learnt and facilitate recitation.137

135 [73] Cf T W Rhys Davids, Vinaya Texts, vol 1, 1881:xxii; R Gombrich 1990a:7; 1990b:24. 136 Allon 1997a:53. See also Harrison, “Mediums and messages,” 2003:123. 137 [74] For example, of the passages which are VR at a primary level, 78.9% consists quantifiable repetition of one kind or another. In total 53.4% of the material which is repetitive on a primary level involves quantifiable repetition on a secondary level.

32 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

4.2.6.9 This study has focused on only one sutta in the Dīgha Nikāya: the Udumbarikā Sīha,nāda Sutta (D 2 5) [4.2.6.1]. But much of the material found in both the repetitive and non-repetitive passages of this Sutta is also encountered elsewhere in the Dīgha Nikāya, which, of course, is significant if a body of suttas such as are contained in the present Dīgha Nikāya as learnt and transmitted by a particular group of monks or nuns. This repetition decreases the uniqueness of the material which is not repeated again within this Sutta, and increases the familiarity of those passages which are. Further, various forms of non-quantifiable repetition are an integral part of all passages, whether these passages are repeated again within this particular sutta or not, whether they are found in other suttas or are unique to this Sutta, whether classified as being repetitive at a primary or secondary level. Passages are built up through the proliferation of similar word elements, units of meaning and structures. Many elements share sound and metrical similarities. Vocatives of address and particles such as atha kho and kho are continually used as markers throughout the Sutta. Certain verbs are repeated in their non- finite forms to resume the following clause.138

4.2.6.10 The wording used to express or depict a given concept, action or event is standardized, and diversity of vocabulary is avoided. And so on. In this way, although we have been able to quantify gross repetition of certain classifications, there are many forms of repetition employed by this class of Pali text which cannot be quantified, yet which must also be considered to facilitate greatly the learning and recitation of this material. Repetition thus thoroughly permeates every dimension of this class of Buddhist literature. The characteristics of the prose portions of Pali canonical sutta texts discussed in this paper show that the authors of this material attempted to minimize differences and maximize similarities. They did this by using a standardized diction (which we have referred to as formulas), by proliferating similar word elements often chosen for their sound and metrical similarities, and by pursuing repetition on a truly large scale, to mention but a few. Of these stylistic features, it is gross repetition which provides the best evidence that these texts were composed as fixed texts which were to be memorized and transmitted verbatim.

4.2.6.11 As previously mentioned, these stylistic features do not prove that this literature was essentially an oral one, for written texts can utilise or mimic characteristics of an earlier oral tradition. Nor do they prove that these texts were conceived as fixed texts. But when combined with such historical factors as accounts of communal recitation, events which required a fixed text, then we are surely on firmer ground. Allon has attempted to show that the early Buddhist sutta texts were, in the words of R Gombrich, “deliberate compositions which were then committed to memory.”139 But he would certainly agree that accounts of what the Buddha is supposed to have said and discourses on his teaching would have been given by the monks and nuns after the Buddha’s death in an improvisatory manner, at times drawing heavily on memorized material, or as R Gethin (1992) has argued, by using lists as a foundation. Such discourses may then have become the basis of later fixed texts. But these accounts and discourses were fundamentally different from the essentially fixed, memorized texts transmitted by the community, however imperfectly. Finally, the Parry-Lord model [4.2.1.1] does not exhaust the oral or literary/performance dimension of oral cultures. In ancient, pre-literate India there was a strong tradition of composing fixed, religious

138 [75] In the above discussion of approach-formulas, eg, upasaṅkami, upasarjcamitvā (“ … he approached. Having approached, … “) and ekam antaṁ nisīdi, ekam antaṁ nisinno (“ … he sat down to one side. Seated to one side … “) were encountered. 139 [76] R Gombrich 1990b:24. http://dharmafarer.org 33 SD 58.1 The Oral Tradition of the Early Buddhists texts which were designed to be memorized and transmitted verbatim, as is the case with the early Bud- dhist oral tradition.

4.2.7 Verbatim repetition (Bhikkhu Analayo)140

4.2.7.1 The oral nature of the early suttas easily impresses itself on the listener or the reader (more so in the former) due to the frequency of verbatim repetition of words, phrases and passages. When treating a particular topic in its positive and negative manifestations, for example, it is standard procedure in the suttas to repeat the same passage with the very same words and formulations used for the positive case, making only the most minimal changes needed in order to adjust these to the negative case. The same procedure becomes even more prominent when a series of different perspectives on a particular topic are explored. Thus, a treatment of, for example, 4 types of persons or modes of acting, will use 4 times nearly the same text in order to achieve its aim. This feature ensures that we understand exactly what the negative points are, and how they are to be corrected or countered by their positive opposite. This is done, as it were, point by point, so that nothing is missed out, as in skimming by way of “intellectual appreciation.” The purpose of such passages is clear: we should not merely “know” them, we should even more than just understand them: we should be transformed by the listening or perhaps the reading.

4.2.7.2 In addition to the frequency of repetition within a single sutta, the early texts also make frequent use of pericopes, that is, formulaic expressions, phrases or passages that depict a recurrent situation or event, and whose purpose is to facilitate memorization141 [4.2.8]. Whether it is a description of how someone approaches the Buddha or of how someone attains liberation, pericopes are employed with a fixed set of phrases and expressions, with only the most minimal changes introduced to adapt these pericopes to the individual occasion. These two features, the repetition of passages within a sutta and the use of pericopes throughout a sutta collection, are responsible for the highly repetitive nature of the early texts.142

140 This section is based on Analayo, “Oral dimension of Pali discourses,” 2007. 141 Cousins notes the “widespread use of mnemonic formulae” (1983:1) as a typical feature of early Buddhist oral literature. Griffith explains that the use of pericopes is “a direct result of the methods by which sacred material was preserved and handed down in the early Buddhist communities; the demands of mnemonic convenience ... meant that the units of tradition ... had to be ... reduced to an easily memorized standard form” (1983:58). Simson compares the function of pericopes in Buddhist prose to the bones and tendons in the human body, in that both provide stability and support for the other parts (1965:47). J D Smith, in an examination of modern oral literature in India, reports the finding that a Rajasthani epic that made frequent use of pericopes (which, according to his description, has the effect that “every battle ... is the same battle, every journey is the same journey, every meeting the same meeting”) was transmitted with considerably greater accuracy than other comparable epics (1987:598). Smith explains that the reason for the employment of pericopes and the resulting greater accuracy “may lie in the fact that the epic is not merely sung for entertainment, but has a religious function,” a reason that would hold true also for the use of pericopes in the oral transmission of the early Buddhist discourses (1977:151). 142 Allon, in his detailed study of these features in Udumbarikā Sīha,nāda S (D 25), Allon comes to the conclusion that over 80% of the text of this discourse involves some form or other of repetition (1997a:359). He concludes that “repetition thus thoroughly permeates every dimension of this class of Buddhist literature.” (1997a:360)

34 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

4.2.7.3 These various oral characteristics of the suttas testify to the importance of verbatim repetition in the early Buddhist oral tradition.143 In this respect, the transmission of early Buddhist texts differs from oral traditions in general, where improvisation is a prominent feature. The performance of oral literature, an epic or narrative type, demands innovation and improvisation from the performer, whose task is to present the main elements of a tale in such a way as to best entertain the audience. This type of oral literature is thus freely re-created every time it is told.144 [4.2.1.1] In contrast, the purpose of the early Buddhist oral tradition was the preservation of instructive and liberating material, for which free improvisation is inappropriate, even counterproductive.145 Moreover, such textual recitation was often undertaken communally by the reciters and audience, which leaves little scope for free improvisation.146 The emphasis on verbatim transmission in the early Buddhist oral tradition can even be detected in some transmission errors, where at times in otherwise closely similar Pali and Sanskrit passages the counterpart to a particular term shows close phonetic similarity, but a significantly different meaning. In such cases, it seems as if the attempt of the reciters to precisely remember has preserved formal aspects, even though the meaning was forgotten or lost.147

4.2.7.4 On account of the nature of the early Buddhist oral tradition, it is only to be expected that such texts orally transmitted over long periods of time, in spite of the various precautions to prevent omissions, variations and errors, they still do occur. Most of the early Buddhist reciters involved in the transmission of the suttas were trained in memorization skills from early youth onwards, as was the case for Vedic reciters.148 While the Vedic reciters were simply trained to memorize the texts, the Buddhist reciters recited the texts in order to understand and practise them for personal liberation. Without early the early conditioning of the virtuosi reciters with eidetic memory, transmission errors nearly unavoidably would creep into these early Buddhist texts. Such variations not only occur amongst the texts transmitted by different Buddhist schools, but can even be found within the texts transmitted by a single school, such as within the Pali texts transmitted by the Theravāda tradition. One type of problem that tend to arise within the Pali tradition involves variations in the use of pericopes. Such pericope variations usually affect those parts of a sutta that were added by the reciters in order to provide a background narrative to the words spoken by the Buddha or his disciples. This is what we will examine next. [4.2.8]

143 Allon explains that “it is surely easier to remember a sequence of words arranged ... according to syllable length,” just as “it is easier to remember two different words when they share sound similarities and have the same metrical pattern” (1997a:252); cf Wynne 2004:108-112. 144 According to Lord, such oral transmission involves “never merely memorizing a fixed entity, but ... ever re- creating a new version of older forms and stories” (1987:71). 145 Bechert points out that oral tradition in India had achieved a remarkable degree of precision (1985:21). Hence, as Graham explains, the “oral transmission of scripture should not be confused with folk oral tradition in which verbatim accuracy is not aspired to” (1987:138). 146 Allon notes that “communal or group recitation or performance requires fixed wording” and would not allow for improvisation (1997b:42). Coward points out that “group listening to check for errors is still an accepted method of verification in rural India today” (1986:300). 147 Simson gives the following examples: brahmujjuggatto - bṛhadṛjugātro; muducittaṁ - mudita,cittaṁ; añña- taro - ajñātavān; sammodi sammodanīyaṁ - sammukhaṁ sammodanīṁ. (1965:137 f) 148 Cf Frauwallner 1956:173-175; Hinüber 1989:67 f. http://dharmafarer.org 35 SD 58.1 The Oral Tradition of the Early Buddhists

4.2.8 Pericope variations149

4.2.8.1 Pericope (from the Latin) is a term for sections of scriptural text, ranging from a unit (phrase), to a paragraph, or even a passage from the Pali canon, such as a single parable, or a single story of some spiritual or textual significance. Often a pericope may appear in the suttas in different context, even with variance in its content. [4.2.7.2] A difference in the use of pericopes can be seen, for example, between a sutta in the Dīgha Nikāya and a sutta in the Aṅguttara Nikāya, two discourses that treat the same event, namely, a visit paid by Ajāta,sattu’s chief minister Vassakāra to the Buddha in order to find out what the Buddha knows or 150 would say about king Ajata,sattu’s plan to attack the Vajjīs. The Dīgha, characterized by its wealth in narratives [3.2.2.1], describes in detail how Vassa,kāra gets his chariot ready, drives with the chariot, and then descends from the chariot to proceed on foot. Its Aṅguttara counterpart, however, does not mention Vassa,kāra’s manner of arrival at all, only simply noting that he approaches the Buddha.

(1) Vassakāra ... assented [to the order given to him] by Ajāta,sattu Vedehi,putta, the king of Māgadha, got the state carriages ready and mounted them, left Rājagaha by state carriage and went towards Mount Vulture Peak. After going as far as the ground was passable for carriages, he descended from the carriage and approached the Blessed One on foot. vassakāro ... rañño māgadhassa ajāta,sattussa vedehi,puttassa paṭissutvā, bhaddāni bhad- dāni yānāni yojapetvā, bhaddaṁ yānaṁ abhirūhitvā,151 bhaddehi bhaddehi yānehi rājagahamhā niyyāsi, yena gijjhakuṭo pabbato tena pāyāsi, yāvatikā yānassa bhūmi yānena gantvā yānā pac- corohitvā pattiko yena bhagavā ten’upasaṅkami. Mahā,parinibbāna Sutta (D 16/2:73,4), SD 9

(2) Vassakaro ... assented [to the order given to him] by Ajata,sattu Vedehi,putta, the king of Magadha, and approached the Blessed One. vassa,kāro ... rañño māgadhassa ajāta,sattussa vedehi,puttassa paṭisuṇitvā,152 yena bhaga- vā ten’upasaṅkami.153 (Sattaka) Vassa,kāra Sutta (A 7.20/4:18.4), SD 72.14

4.2.8.2 Another case where the records of the same event differ in the detail in which they depict how someone approaches the Buddha can be found between a total of 4 suttas that describe the fam- 154 ous last meeting between Māra and the Buddha, in which Māra asked the Buddha to pass away. While the Dīgha and the Udāna versions report that Māra approached the Buddha, stood at one side and then addressed the Buddha, the Saṁyutta version of the same event does not mention that he stood at one side, but only records that he approached the Buddha. The Aṅguttara version does not record any approach at all.

149 This section is based on Analayo 2007:10-14. 150 Allon 1997a:39. 151 Be Se yojetvā, bhaddaṁ bhaddaṁ. 152 Be paṭissutvā; Ce paṭissutvā utthāy’āsanā. 153 Allon notes that a description of how someone approaches by chariot can, however, be found elsewhere in Aṅguttara, eg, A 5.50/3:59,27 (king Muṇḍa approaches the monk Nārada); A 8.12/4:181,23 (general Sīha approaches the Buddha); and A 10.30/5:65,9 (king Pasenadi approaches the Buddha); though the description given in these suttas is shorter than the “chariot approach” pericope employed in Dīgha. (1997a:39) 154 Allon 1997a:62.

36 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

(1) Not long after venerable Ānanda had left, Māra the Evil One approached the Blessed One; having approached, he stood on one side; standing on one side, Māra the Evil One said this to the Blessed One ... māro pāpima acira,pakkante āyasmante ānande yena bhagavā ten’upasaṅkami, upasaṅ- kamitva ekam antaṁ aṭṭhāsi; ekam antaṁ ṭhito kho maro pāpimā bhagavantaṁ etad avoca ... Mahā,parinibbāna Sutta (D 16/2:104,12 = U6.1/63,13), SD 9

(2) Not long after venerable Ānanda had left, Māra the Evil One approached the Blessed One; having approached, he said this ... māro pāpimā acira,pakkante āyasmante ānande155 yena bhagava ten’upasaṅkami,156 upa- saṅkamitvā157 etad avoca ... . Cetiya Sutta (S 51.10/5:260,25)

(3) Not long after venerable Ānanda had left, Māra the Evil One said this to the Blessed One ... māro pāpimā acira,pakkante āyasmante ānande bhagavantam etad avoca. Bhūmi,cāla Sutta (A 8.70/4:310,11)

4.2.8.3 Pericopes also differ when it comes to describing the respectful attitude with which someone listens to an after-meal sermon given by the Buddha or by a monk. For such occasions, the Dīgha, the Majjhima, the Udāna, and the Sutta Nipāta employ a pericope that describes how the listener(s) take(s) a 158 low seat, an obvious expression of respect. 159 Similar situations in the Vinaya and in the Aṅguttara, however, do not mention a low seat. This difference is particularly notable in the case of a meal given by prince Bodhi, as the same meal is recorded in the Majjhima and the Vinaya, so that, in this case, the same event is described once with taking a low seat and once without doing so.

(1) When the Blessed One had eaten and had removed [his] hands from the bowl, prince Bodhi took a low seat and sat down on one side. bodhi rāja,kumāro ... bhagavantaṁ bhuttaviṁ onīta,patta,pāniṁ aññataraṁ nīcaṁ āsanaṁ gahetvā ekam antaṁ nisīdi. (M 85/2:93,9)

(2) When the Blessed One had eaten and had removed [his] hands from the bowl, Prince Bodhi sat down on one side. bodhi raja,kumāro ... bhagavantaṁ bhuttaviṁ onīta,patta,pāṇiṁ ekam antaṁ nisīdi. (V 2:128,36)

155 Ce omits acīra,pakkante āyasmante ānande. 156 Se adds (ekam antaṁ aṭṭhā. Ekam antaṁ ṭhito kho māro pāpimā) in brackets. 157 Be Se add bhagavantaṁ. 158 The pericope of “taking a low seat,” aññataraṁ nīcaṁ āsanaṁ gahetvā, leads from the pericope that describes the giving of a meal to a sermon, eg, in D 3/1:109,36 (for further reference and a discussion of this variation, cf Allon 1997a:122). The same pericope can also be found regularly in Madhyama Āgama, cf, e.g, MĀ 132 (T1.625b17): 取一小床, which is also found in Tibetan counterpart at D ’dul ba kha 105b3: stan ches dma’ ba zhig blangs te, but the whole episode is missing from Pali version at M 82/2:64,23. Skt occurrences, eg: nīcataram āsanaṁ gṛhītvā in Dutt 1984:65,15, being a counterpart to Sn 3.7/111,9: aññataraṁ nīcaṁ āsanaṁ gahetvā; or (nīcata)[r](a)[k](a)m-asan- aṁ gṛhītvā in Mahāparinirvāṇa Sūtra fragment S 360 folio 187 V5 in Waldschmidt 1950:26, being a counterpart to D 16/2:126,26, where the low seat is not mentioned; or nīcataramṁ āsanaṁ gṛhitvā in Saṅghabheda,vastu in Gnoli 1977: 45,14, being a counterpart to V 1:18,9, where the low seat is not mentioned. 159 Instead of the phrase “taking a low seat” after the pericope that describes the giving of a meal, only the phrase “sat down at one side,” ekam antaṁ nisīdi, leads up to a sermon, eg, in A 4.57/2:63,4 (for further ref: Allon 1997a: 123). http://dharmafarer.org 37 SD 58.1 The Oral Tradition of the Early Buddhists

4.2.8.4 The application of a pericope can at times result in inconsistencies within the same sutta. An example is the pericope that describes how the Buddha or a monk gets ready to go on almsround. Since a monastic has to take his food before noon, such preparations are usually made in the early morning, so that this pericope describes how “in the morning” the Buddha or a monk dresses and takes his bowl 160 and robe, and approaches the village or town. The frequency of this pericope has caused it to be also applied to a passage in the Vinaya and the Udāna where it does not seem to fit its context. This passage records how the Buddha is travelling and arrives in a particular place where he is invited to come to the local hall. The villagers then approach the same hall and listen to a discourse by the Buddha that goes on well into the night. Even though the circumstances make it clear that the invitation to come to the local hall must have taken place in the late afternoon or evening, the Vinaya and the Udāna nevertheless report that it is “in the morning” that the Buddha responds to the invitation by dressing and taking bowl and robe to approach the local hall.161

4.2.8.5 The relatively circumstantial differences noted so far may seem negligible, since they do not affect essential matters. Not all such errors, however, are of such circumstantial character. A somewhat more significant variation in the use of the pericopes employed at the conclusion of a discourse can be found between the Saṁyutta and the Sutta Nipāta versions of the Kasi Bhāra,dvāja Sutta. These 2 suttas treat the same event but differ in their conclusion, as according to the Saṁyutta account Kasi Bhāradvāja takes refuge and declares himself to be a lay follower, while according to the Sutta Nipāta version he takes refuge, requests for ordination and becomes an arhat.

(1) I go for refuge to master Gotama, to the Dhamma and to the community of bhikshus, may master Gotama remember me as a lay follower who from today on has gone for refuge for life. 162 esâhaṁ bhavantaṁ gotamaṁ saraṇaṁ gacchāmi dhammañ ca bhikkhu,saṅghañ ca, upāsakaṁ maṁ bhavaṁ gotamo dhāretu ajjatagge pāṇupetaṁ saraṇaṁ gatan’ti (S 7.11/1:173 reads dharetu; S2 197/1:372,20)

(2) I go for refuge to master Gotama, to the Dhamma and to the community of bhikshus, may I receive the going-forth in the presence of master Gotama and the ordination ... and venerable Bhāradvāja became one of the arhats.163 164 esâhaṁ bhavantaṁ gotamaṁ saraṇaṁ gacchāmi dhammañ ca bhikkhu,saṅghañ ca, labheyyâhaṁ bhoto gotamassa santike pabbajjaṁ labheyyaṁ upasampadan’ti ... aññataro ca 165 kho panâyasmā bhāradvājo arahataṁ ahosi. (Sn 1.4/15,23)

160 Eg M 5/1:31,29: pubbaṇha,samayaṁ nivasetvā patta,civaraṁ ādāya. 161 V 1:227,10 = U 8.6 /86,13: bhagavā pubbaṇha,samayaṁ nivasetvā patta,civaraṁ ādāya saddhiṁ bhikkhu,- saṅghena yena āvasathâgāraṁ ten’upasaṅkami, followed by describing that the laity heard a sutta from the Buddha and were then dismissed, bhagavā ... upāsake bahud eva rattiṁ dhammiyā kathāya sandassetvā samā- dapetvā samuṭṭejetvā sampahaṁsetvā uyyojesi, where the reference to bahud eva rattiṁ makes it clear that the discourse was given at night time, so that the earlier reference to “the morning,” pubbaṇha,samayaṁ, does not fit the context. Allon notes this error and also draws attention to instances where the pericope is properly adjusted to its context eg in M 53/1:354,12 or in S 35.202/4:183,16, which introduce a similar situation only with nivasetvā patta,civaraṁ ādāya, without the specification pubbaṇha,samayaṁ (1997a:141). 162 Following Ce bhavantaṁ; Ee bhagavantaṁ; Be Se abbreviate. 163 Both the Chinese parallels SĀ 98 (T2.27b26), SA2 264 (T2.466c10) and SA3 1 (T2.493b8) agree with Sn 1.4 in saying that he goes forth and becomes an arhat.

38 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

4.2.8.6 In regard to variations in the use of pericopes, it is also of interest to compare their use in Pali suttas to the usage in their parallels in the Chinese Āgamas. Taking as an example the Madhyama Āgama preserved in Chinese, sutras in this collection regularly describe how a monk fans the Buddha,166 a situation noted only rarely in the Pali suttas, the Majjhima Nikāya.167 The Madhyama Āgama sutras also often mention the sitting-mat,168 one of the allowed accessories or gears of a monk, while the Pali suttas refers to it only on very few occasions.169 The two collections also differ in their descriptions of how listeners will express their appreciation of the teachings, since whereas in a Majjhima Nikāya discourse they exclaim “wonderful, wonderful”, in a Madhyama Āgama discourse they rather inform the Buddha: “I understood, I realized”.170 Again, when someone asks the Buddha or a monk a question in the Madhyama Āgama, the actual question will be preceded by a request to be given permission to put a question,171 a pericope found only rarely in the Majjhima Nikāya.172

4.2.8.7 Analayo raises an interesting point: “Another standard pericope in the Madhyama Āgama describes how a visitor or a monk will depart from the presence of the Buddha by performing three circumambulations, a circumstance not mentioned in the Majjhima Nikāya discourses.” (2007:13). Analayo then points out (13 n46) that the triple circumambulation is mentioned at MĀ 132 (T1.26.623b23) 繞三匝 rào sān zā, and its parallel M 82/2:56,22: padakkhiṇaṁ karoti and its various forms. He adds that a reference to it is found in the Mahā,parinibbāna Sutta (D 16/2:163,27), when Mahā Kassapa performs the triple circumambulations around the Buddha’s funeral pyre (citaka).173 There are at least 2 reasons for the rarity of mention of the triple circumambulation in the suttas:

(1) It is never done to a living person (not in the suttas). When taking leave of the Buddha or a respected person (a monk, a teacher, etc), one would depart “keeping the right side” (padakkhiṇa) to him. This gesture apparently evokes the idea of treating the sacred person like the sun, and one thus moves “sunwise” (today we would say “clockwise”) away from that person. This is only done once, as it would be inconvenient, even distracting, for the honoured person and his audience, for this gesture of respect to be a prolonged ritual done thrice!

(2) The Chinese Āgama parallels to the suttas often mention the triple circumambulation due to the historical fact that it was already a popular practice by the time the translations were done in China

164 Se adds upāsakaṁ maṁ bhavaṁ gotamo dhāretu ajjatagge pāṇupetaṁ saraṇaṁ gataṁ before labheyyâhaṁ etc. 165 Be omits kho, Se omits ca. 166 Eg MĀ 33 (T1.26.474a19) 執拂侍佛 zhí fú shì fó. 167 M 12/1:83,20 and M 74/1:501,1 state that a monk is fanning the Buddha. 168 Eg MĀ 9 (T1.430b10): 尼師檀 ní shī tán (with a 宋 Sòng, 元 Yuán and 明 Míng variant reading 尼師壇 ní shī tán); cf Thich Minh Chau 1991:29. 169 M 24/1:147,5 and M 147/3:277,30. 170 Eg M 7/1:39,27: abhikkantaṁ ... abhikkantaṁ, and its counterpart in MĀ 93 (T1.576a10): 我已知。我已解。 wǒ yǐ zhī, wǒ yǐ jiě 171 [48] Eg MĀ 29 at T 1.461b27-28: 欲有所問，聽我問耶。 yù yǒu suǒ wèn, tìng wǒ wèn ye. 172 M 35/1:229,35, M 109/3:15,23, M 144/3:264,30. Notably, though none of these 3 Pali suttas has a parallel in Madhyama Āgama, each has a parallel in Saṁyukta Āgama, but in each case this pericope is not found, cf SĀ 110 (T2.35c11 parallel to M 35); SĀ 58 (T2.14b17 parallel to M 109); and SĀ 1266 (T2.347c23 parallel to M 144). A Tib version of this pericope parallels M 90, D ’dul ba kha 88a1, where this particular pericope is also found in Madhya- ma Āgama parallel to the same text, MĀ 212 (T1.793b15), but not in the Pali version. 173 Part of this pericope has also been preserved in a Skt fragment parallel to an occurrence of this pericope in MĀ 161 (T1.686a18); cf SHT V 1148 R4 in Waldschmidt 1985 147: triprada(kṣinīkṛtvā). http://dharmafarer.org 39 SD 58.1 The Oral Tradition of the Early Buddhists

(from the Han dynasty, 1-221 CE, onwards), and it was thus mentioned so often in the Chinese translations. Note that the case quoted by Analayo above refers to Mahā Kassapa honouring the Buddha’s remains (which is regarded as relic, that is, a shrine, cetiya). After the Buddha’s passing (perhaps, following Mahā Kassapa’s initiative), the triple ambulation naturally evolved into a ritual of respect to a shrine (cetiya) and later, a stupa (thūpa).

It should also be noted that in the suttas, we only see the disjunct phrase, tikkhattuṁ … padakkhiṇ- aṁ katvā, “having performed the rightwise ambulation” (D 16/2:163,27), and its various grammatical forms. The noun in this phrase’s hiatus is citakaṁ. “a funeral pyre (for a body), a tumulus (for bones or relics).” This same phrase recurs in the Vinaya, where a story of the past relates Brahmadatta, king of Kāsī, seeing prince Dīgh’āvu thrice circumambulating his father’s funeral pyre (citaka) with palms in lotus gesture (pañjalikaṁ tikkhatuṁ citakaṁ padakkhiṇaṁ karontaṁ). The prince is honouring the remains of his dead father.174 The Netti-p,pakaraṇa quotes an untraced “sutta on past karma” (vāsanā,bhāgiyaṁ suttaṁ): “Keep- ing rightwise, | the kinsman of the sun’s head, ǁ he revolved around (it) thrice, | (then) vanished away on the brow” (padakkhiṇaṁ karitvāna | sīse ādicca,bandhuno ǁ tikkhattuṁ parivaṭṭetvā | muddhan an- taradhāyatha,175 Nett 138,20 f). In the Sakk’udāna Sutta (U 29), Sakra, departing rightwise from Mahā Kassapa, rises into the air and thrice jubilantly utters the udāna: “O what a gift! A supreme giving has been well placed in Kassapa!” (aho dānaṁ parama,dānaṁ kassape suppatiṭṭhitaṁ!) He only moves rightwise (no triple ambulation).176

4.2.8.8 A wild-card search of the Chaṭṭha Saṅgāyanā CD using “padakkhiṇ*”177 yielded the following hits of the triple circumambulation being done to the Buddha by various people (unless otherwise stated):

Totals Dīgha 1:85 89178 125 14 225 2:40 47 48 84 86 96 97 104 127 132 133 163x2179 204 252 3:205x2 206x2 208 209 26 times Majjhima 1:146 147 169 252 304180 342 354 376181 380 393182 393 457 496 2:48 50 56x2 61 93 102 117183 124 133 135 145 208 3:247 262 263 269 276x2 277x2 34 times Saṁyutta 1:1 18 46 49 50 56x2 120 138 150 151x2 153 3:2 6 36 74 92 121 4:38 63 64 76 183 184 (284 285 289)184 324 377185 380 5:144 166x2 187 218 260 37 times

174 A verse in Paduma Bv reads vanditvā sirasā | katvāna taṁ padakkhiṇaṁ ǁ tikkhattuṁ abhinanditva | sattāhaṁ jinam-upaṭṭhahiṁ, “With my head, I honoured his feet | circumambulated him ǁ roared thrice (loudly) | and served the Conqueror for a week” (B 9.11). 175 Muddhan’antaradhāyatha is obscure: the tr follows Nett:Ñ 185. 176 U 3.7/27/29 (SD 71.3). 177 The hits are actually higher when we also collate other appearances of padakkhiṇ*, eg, padakkhiṇā, padak- khiṇe etc, and incl the hits in the Khuddaka Nikāya books. 178 Ambaṭṭha to Pokkharasāti (D 3,1.6/1:89), SD 21.3. 179 The elder Kassapa honouring the Buddha’s remains, and the 500 monks with him do the same (D 16,6.22/- 2:163), SD 9. 180 Layman Visākha to nun Dhammadinnā (M 44,30/1:304), SD 40a.9. 181 Layman Upāli to Nirgrantha Nātaputta (M 56,9/1:376), SD 27.1. 182 Prince Abhaya to Nigaṇṭha Nāta,putta (M 58,4+5/1:393), SD 7.12. 183 King Pasenadi to Ānanda (M 88,20/2:117), SD 49.12. 184 The houselord Citta to monks (S 41.2/4:284 f), SD 65.10, & (S 41.4/4:289), SD 27.2. 185 King Pasenadi to the nun Khemā (S 44.1/4:377), SD 63.6.

40 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

Aṅguttara 1:225 277 294x3186 2:21 182 249 3:37 59187 70 123188 123 (198 199x2)189 330 331x3 359x2 423 424 4:28 29x2 30 59 75x2 187 220 274 301 310 356 309x2 374 5:58190 69 170 172x2 189 192 320 48 times Udāna 16 25 30191 35 38 49 58 63 81 86 87 92 12 times Sutta Nipāta pp124 125x2 Sn 1010192 4 times

The Dīgha yields a total of 26 occasions of performance of the triple circumambulation, all of which, except 1, are directed to the Buddha. The exception is in the Ambaṭṭha Sutta (D 3), when the brahmin youth Ambaṭṭha honours his teacher, Pokkharasāti.

The Majjhima yields 34 hits, of which 30 times are shown to the Buddha. The exceptions are in: Cūḷa Vedalla Sutta (M 44/1:304) the layman Visākha to the nun Dhammadinnā; Upāli Sutta (M 56/1:376) the layman Upāli to Nirgrantha Nātaputta; Abhaya Rāja,kumara S (M 58/1:393) prince Abhaya to Nirgrantha Nātaputta; later, to the Buddha, too; Bāhitika Sutta (M 88/2:117) king Pasenadi to the elder Ānanda.

The Saṁyutta yields 37 hits, of which 33 times refer to the Buddha. The exceptions are in: Isi,datta Sutta 1 (S 41.2/4:284 f) the houselord Citta to the monks; Mahaka Sutta (S 41.3/4:289) the houselord Citta to the monks; Khemā Therī Sutta (S 44.1/4:477) king Pasenadi to the nun Khemā.

The Aṅguttara yields the most, 48 times, of which 39 refers to the Buddha. The exceptions are in: Pubb’aṇha Sutta (A 3.155/1:294) padakkhiṇa as meaning “auspicious, good” in terms of deeds; Nārada Sutta (A 5.50/3:59) royal treasurer to the elder Nārada; Kakudha Sutta (A 5.100/3:123) devaputra to the elder Moggallāna; Padakkhiṇa as faith and respect (A 5.167/3:198, 199x2); Mahā Pañhā Sutta 2 (A 10.28/5:58) lay followers of Kajaṅgalā to the nun of Kajaṅgalā. [5.2.9.3]

The Udāna gives 12 occasions of which 11 refers to the Buddha; the exception is in:

Sakka’udāna Sutta (U 3.7/27/29) Sakra to Mahā Kassapa: [4.2.8.7].

The Sutta Nipāta gives 4 hits, all referring to the Buddha, except for Sn 1010, which refers to the 16 youths saluting their teacher Bāvari (Sn 1010).

4.2.8.9 Other pericopes, found in the Majjhima Nikāya, are absent from the Madhyama Āgama. One example is the pericope used regularly at the start of a Majjhima sutta, in which the Buddha addresses the audience with “Monks!” and they reply “Bhante!” (venerable sir), after which the Buddha announces his topic and proceeds to teach. Such an exchange is not found in the Madhyama Āgama discourses.193 According to Analayo, a close inspection shows that this pericope does not fit too well with the remain-

186 Padakkhiṇa as “auspicious, good” regarding actions of body, speech and mind: Pubb’aṇha S (A 3.155/1:294). 187 Royal treasurer Piyaka to the elder Nārada (A 5.50/3:59). 188 Devaputra Kakudha to the elder Moggallāna (A 5.100/3:123), SD 70.6. 189 Padakkhiṇa as gesture of faith and respect (A 5.167/3:198, 199x2), SD 88.1. 190 Lay followers of Kajaṅgalā to the nun of Kajaṅgalā (A 10.28/5:58), SD 85.16. 191 Sakra to Mahā Kassapa (U 3.7/27/29), SD 71.3: [4.2.8.7]. 192 The 16 youths to their teacher Bāvari (verse) (Sn 1010/194). 193 Eg in M 1/1:1,3: Bhikkhavo’ti. bhadante’ti te bhikkhu bhagavato paccassosuṁ; on the use of this pericope in Dīgha and the Majjhima, cf Manné 1990:33; on its relative lateness: Meisig 1987a:25. http://dharmafarer.org 41 SD 58.1 The Oral Tradition of the Early Buddhists der of the Pali sutta in which it occurs, where the vocative “Monks!” as bhikkhavo194 is used, differing from the vocative form, bhikkhave, used in the remainder of the discourse.195 Similarly, the first vocative bhadante (venerable sir), used by the monks, is not the same as the vocative bhante, used by them elsewhere in the discourse.196 Since there would be no reason for starting with one particular vocative and then switching to another type of vocative, argues Analayo, this difference suggests that this pericope may have been added during oral transmission. Though this pericope is not found in the Madhyama Āgama, it does occur in an individual translation, that is, a discourse translated individually into Chinese. Notably, this discourse comes from a Madhyama Āgama collection that is no longer extant, says Analayo.197 Such problems may seem abstruse, even trivial, to the general reader, but they fascinate the specialist who have some profound understanding of how the early Buddhist oral tradition grew and is able to organize itself into some form so that the canonical texts are fully and safely handed down to us to this day, and for posterity. This is what we will look into next.

5 Genre, āgama, mātikā, aṅga

5.1 GENRES

5.1.1 Evolution of the early oral tradition

5.1.1.1 The evolution of the early Buddhist oral tradition from its beginning with the Buddha’s discourses down to the sutta legacy that we are blessed with today lies in these 3 terms: āgama, aṅga, ni- kāya. In simple terms, we can say that the oral teachings began with āgama (singular), the Dharma as a practical living whole, the teaching, practising and realizing of the Dharma. This idea of knowing and seeing the Dharma for oneself evolved into aṅga, limbs or classes of Buddha Dharma. In the final stage, we have the nikāya (plural), that is, the Tipiṭaka as we have it today. [6]

5.1.1.2 These are the 3 basic stages in the evolution of the early Buddhist oral tradition, that is, as realization, as theory, and as texts. In the earliest days of the ministry, those who followed the Buddha, as a rule, quickly and easily attaining awakening for themselves. Their practice and experience of awakening, their realization, is called āgama [5.2]. In due course, the awakened taught the teaching to more of the unawakened, for whom the Dhamma remained, for a while at least, as theoretical limbs or classes of teachings—this is called aṅga [5.4]. And finally, for our benefit, these records of realization and their teachings are put together as standard texts, as nikāya [6], so that we have access to the early Bud- dhist teachings even today, to guide us in our understanding of the Buddha Dharma so that we are able to practise it just as it is done in the early days of the Buddha himself or intended so by him.

194 While bhikkhavo, “Bhikshus!” is voc pl is the “Western” form (D 2:120,21*; V 1:19,30 = 20,28), bhikkhave is the “Eastern” form (Oberlies §5.2; p108). See Geiger, Pali Grammar, §82.5; Hinūber, Das ältere Mittelindisch im Überblick, 2001, §332; Oberlies, Pali Grammar, 2019:282 (index 2035). 195 M 1/1:1,5 continues with bhikkhave. On the voc bhikkhave, cf Bechert 1991:11; Lüders 1954:13. 196 M 1/1:1,7 continues with bhante. 197 T48 (T1.837c25-26): “The Buddha said: ‘Monks!’; the monks replied: ‘Yes, indeed!’; the monks listened to the Buddha; the Buddha said ... ,” 佛告諸比丘,比丘應曰唯然,比丘從佛聽佛說 (fó gào zhū bǐqiū, bǐqiū yìng yuē wéi rán, bǐqiū cóng fó tīng, fó shuō). According to the introductory remark in T48 (T1.837c22), this discourse comes from the Madhyama Āgama, 出中阿含 (chū zhōng ā hán).

42 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

5.1.2 The 7 genres in early Buddhism

5.1.2.0 In this discussion, it helps to understand and use the term genre [zhah.nr], the French term for a type, species or class of composition. A literary or textual genre is a recognizable and established category of work, oral or written, in the case of Pali texts, employing such common conventions. Among them are epics, romances, lyrical passages, humour, satires, tragedy, melodrama and elegies.198

5.1.2.1 Epic is traditionally the largest and highest of literary form, a genre that defines or reflects a whole culture or practice as culture in the case of early Buddhism. An example of a sutta epic is the Mahā Sudassana Sutta (D 17). This epic describes in glorious sensual detail all of the 6 sense-bases, what really is a meditative experience, especially that of dhyana (jhāna). SD 36.12.

5.1.2.2 Romance, in early Buddhist textual literature refers to usually a legendary story in prose or verse about an idealized character, especially the Bodhisattva, the Buddha’s past lives (such as those in the Jātaka, canonical and commentarial), the live of past buddhas, and even some aspects of the Buddha’s life, especially those involving extraordinary or miraculous states and events. In significant ways, such a romance serves to relate some kind of psychological reality rather than the entertaining realism. The traditional (non-canonical) story of the Bodhisattva’s life is a “quest romance,” whose theme is a hero’s movement away from his parents, status, pleasures and home, in quest of that which is beyond decay, disease and death. The narrative relates, as it were, an external quest that we should internalize as a commitment to renunciation in imitation of the Bodhisattva. The most interesting of ancient romances must surely be that of the Mahosadha Jātaka (J 546), also called the Maha Ummagga, “the great tunnel,” after the climax of the beautiful romance of the Bodhi- sattva instructing, yet delighting, us with his remarkable wit and wisdom.199

5.1.2.3 Lyric is technically a pure or true poetic work, usually shorter than a narrative poem (like a canonical Jātaka story); hence, it is properly removed from the more worldly and rhetorical. A lyrical poem often evokes joy, especially that of renunciation, and may be chanted by monastics or sung by the laity. A good example of this is the Dhaniya Sutta (Sn 1.2), SD 50.20.

5.1.2.4 Humour, in the early Buddhist texts, refer to the 4 basic states, or better, conditions, that comprise and pervade all material existence or form (rūpa). In the western sense, especially since mediaeval times, “humours” refer to bodily fluids to which mediaeval medicine attributed the various types of human temperaments, according to which predominates in our body. The preponderance of blood makes one “sanguine”; an excess of phlegm makes one “phlegmatic”; too much choler (yellow bile) makes one “choleric; too much black bile, “melancholic.” Although the early Buddhist teachings do have a similar idea about these humours, they relate more to the impermanent nature of the body and all physical existence. Instead of the western mediaeval idea of the 4 humours (which could have derived from India, anyway), early Buddhism gives an early conception of physical health based on the 4 elements [5.2.6.7] and also karma.200 In Buddhist humour, we often see the human infected by some subhuman, even mechanical, predi- lection, so that we are enslaved, as it were, to routine, instead of being free of it through renunciation; or, failing to see our own human or divine state, and stooping down, enticed, for lesser or baser things; or, a

198 For scholarly defs of these terms, see eg Chris Baldick, The Concise Oxford Dictionary of Literary Terms, 1991. 199 J 546/4:329-478. See Winternitz 1933:137-139. 200 On early Indian Buddhist medicine, see SD 5.6 (1). http://dharmafarer.org 43 SD 58.1 The Oral Tradition of the Early Buddhists monastic, forgetting his state, indulges in worldly ways (such as eating unmindfully). This last would be described as a “comedy of manners.” Such humour pervades much of the Aggañña Sutta (D 27), SD 2.19. 5.1.2.5 Satire is essentially a “stew” (Latin satura), a mishmash or jumble, and incongruity, a juxtaposition of the high and low. Unlike comedy, satire relies on memorable, often grotesque images of humiliation,201 ridicule,202 sexuality,203 even of physical or criminal violence.204 The brahmins are often the target of Buddhist satire on account of their monomaniacal, narcissistic, claims of biological purity and religious supremacy. Satire, then, is often a grosser, more directed, even personal, form of humour. We see the bad monastic satirized in the Aggañña Sutta (D 27), and the brahmin satirized in the Te,vijja Sutta (D 13), SD 1.8.

5.1.2.6 Tragedy is a serious story relating the death or downfall of a key character, usually the protagonist. Since early Buddhism teaches rebirth, the tragedy only extends as far as this life. When the protagonist is a virtuous person, he gains a happier rebirth or even attains the path or awakening itself. The idea behind a Buddhist tragedy is that bad never wins, that good always triumphs in the end. The best known of Buddhist tragedies must surely be the story of Sāmā,vatī, the 3rd chief queen of king Udena, as related in the Sāmā,vatī Vatthu (DhA 2.1). Māgandiyā, Udena’s 2nd chief queen, nurses a deep grudge against the Buddha for jilting her.205 When she learns that Sāmāvatī is his fervent devotee, she plots her tragic death along with her 500 handmaidens. The plot was uncovered and she was herself painfully put to death by Udena.206 Accounts of monks’ suicides may also be categorized as tragedies. Hence, we can include here, for example, the stories of the last days of the elder Vakkali, as recorded in the Vakkali Sutta (S 22.87),207 and of the elder Godhika of the Godhika Sutta (S 4.23).208 Both monks killed themselves so that they die as arhats (by not falling back from that moment of mental release). Dying as arhats, Māra is unable to detect their consciousness (they are not reborn).

5.1.2.7 Elegy is technically an elaborately formal poem (all early Buddhist poems can be chanted; hence, also sung), highlighting or lamenting the death of a central character, or reflecting seriously on a solemn subject. Such an elegy would be the Amba,pālī Therī,gāthā, where she movingly and graphically reflects on the impermanence that has overtaken her erstwhile beauty (Thī 252-270). This genre, by its very nature, would include a wide range of teachings on the reflection on death, such as that in the Vijaya Sutta (Sn 1.11), given in connection with the death of Rajagaha’s courtesan,

201 Lāl’udāyī,thera V 3 (DhA 18.4) relates how the boastful Udāyī, claims to be able to speak Dharma like the 2 chief disciples. On failing to do so, the crowd routed him (DhA 18.4/3:344-347), SD 50.3(2). 202 Sūkara J (J 153) relates how a foolish boar covered in dung stops a lion from killing him for his meal. This is told in connection with the arrogantly dull elder, Udāyī, for belittling the Buddha’s 2 chief disciples. (J 153/2:912), SD 50.3(3). 203 A most ribald Chaucerian tale is Kacchapa J (J 273), where a mischievous monkey inserts his organ into a sleeping tortoise mouth, who upon waking snapped his mouth shut. The hapless monkey seeks the Bodhisattva’s help, who does so by making the tortoise laugh! The story is told in reference to the quarrelsome ministers of the king of Kosala. (J 273/2:359-361) 204 Makasa J (J 44) relates how when a bald old carpenter annoyed by a mosquito buzzing on his head, asks his foolish son to rid him of it. The son uses a sharp axe and strikes at the mosquito, splitting the father’s head! The story’s moral is that a sensible enemy is better than a foolish friend. (J 44/1:246-248) 205 DhA 2.1(5)/1:199-203. 206 DhA 2.1(3-6)/1:187-225. 207 S 22.87/3:119-124 (SD 8.8). 208 S 4.23/1:120-122 (SD 61.16).

44 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

Sirimā, for the benefit of the lovesick monk who was infatuated sick with her.209 On a simpler level of prac tice, we have the reflection of impermanence, such as that given in the Pheṇa,piṇḍa Sutta (S 22.95).210 5.1.2.8 Genres only describe the type or theme of the sutta or story: this is a literary classification of the early Buddhist texts. The suttas, for example, exhibit many, if not all, of these genres. These suttas are in both prose and verse, and often have both mixed together. Much of these texts simply elaborate the Buddha’s teaching in a straightforward didactic manner: they simply explain the teaching. Some of these teachings record inspired and joyful presentations of the teaching, such as the Udāna. And there are more stories about the Buddha and his immediate disciples. To accommodate such classes of teachings, there is another kind of early Buddhist taxonomy, known as āgama, which we will now turn to.

5.2 ĀGAMA: ALL THE TEACHINGS IN PRACTICE

5.2.1 Āgama as authenticity and tradition

5.2.1.1 Āgama has the basic meaning of received tradition or a canonical text (oral tradition) as the basis for scriptural authenticity and authority, that is, the Buddha-word (buddha,vacana). We can translate āgama here as “text” in the sense of the spiritual experience and awakening preserved in a canon of teachings from those who are awakened or at least attained the path as a streamwinner (as in the case of the elder Ānanda in the Buddha’s life-time). This is the way the term is used in the suttas.211 A clear example of āgatâgama in the sense of “an attained disciple” is that of an unnamed “nun of Kajaṅgalikā.” When she is approached by lay followers to explain a teaching recently given by the Bud- dha, she replies that she has not heard it, but will answer according to “what it means to me.” When the lay followers reported her reply to the Buddha, he praises her for being very wise. [5.2.9.3]

5.2.1.2 The term āgama, as it is used in the suttas, occurs in the singular, as a collective noun, in the sense of spiritual experience or an idea of it [5.2.1.1]. It appears in this sense in the phrase, āgatâgama, “who has mastered attainment (sg)” or “who has attained mastery,” that is, one learned in the true teaching, what is in accordance with the Dharma. This phrase often forms part of the āgatâgama pericope, a description of attained and learned elders who are said to be bahu-s,suta āgatâgama dhamma,dhara vinaya,dhara mātikā,dhara (both in the singular with -o ending,212 and the plural with -ā ending).213 He is (They are) learned, deeply learned, mastered attainment, Dharma-expert, Vinaya-expert, Matrix-expert [4.1.2.5]. The Vinaya pericope adds that they are: “wise, expert, intelligent, conscientious, concerned, devoted to the training” (paṇḍito vyatto medhavī lajjī kukkuccako sikkhā,kāmo).214 They are those who should be approached so that we properly receive Dharma teachings or clarifi- cations about the Dharma.215 Alternatively, the passage in question may take up the need for such learned elders to give teachings.216

209 Sn 1.11/193-206/34 f (SD 3.8(6)). 210 S 22.95/3:140-143 (SD 17.12). On genre in the early Buddhist literature, further see Wynne 2004:113-116. 211 Barua notes that “in the Pāli discourses, ascribed to the Buddha himself, the expression Āgama is often met with, no doubt in the sense of a floating body of Buddhist literary traditions.” (1923:359) 212 As āgatâgamo (sg): A 2:127,30, 170,1+13; V 1:118,28, 119,22 = 337,12, 127,30, 128,13, 338,19+35, 2:8,28, 55,7+13+19, 58x3, 98,20, 299,16+19, 300,10, 301,30, 4:158,6+18 upāsako +; Miln 19,14, 21,16; J 6:475,30*. 213 As āgatâgamā (pl): D 2:125,6 = S 2:169,18; M 1:221,21 = A 5:349,16; A 1:117,28, 2:147,31, 148,24, 169,18, 170, 3:179,2, 180,7, 361,23, 362,18, 5:15,31, 352,10; V 1:338,35, 339,7+34, 2:98,31. 214 V 1:119,22, 127,30, 337,12, 338,20, 2:8,28, 55,29, 98,5 299,16, 4:158,18 of a layman. 215 A 4.160/2:147,29, 5.156/3:179,2. http://dharmafarer.org 45 SD 58.1 The Oral Tradition of the Early Buddhists

5.2.1.3 The term āgatâgama occurs most often in the Aṅguttara Nikāya and Vinaya, of which no parallels have been preserved outside. The āgatâgama pericope also occurs in the Mahā Gopālaka Sutta (M 33),217 3 parallels found in the Chinese that mention such elders. They, however, are only said to be “learned,” without the other qualities—there is no mention of āgatâgama, no āgatâgama pericope—as in the Mahā Gopālaka Sutta.218 By then, either the pericope had been forgotten or the term āgama had taken on a new meaning. [5.2.2] The Mahā,parinibbāna Sutta (D 16) mentions an elder or several who have “mastered attainment (āgama)” as amongst those who should be approached as the 3rd “great reference (mahā’padesa) for verification or authentication that “this is the Dharma, this is the Vinaya, this is the Teacher’s teaching.”219 A Sanskrit fragment parallel has preserved only a reference to the elders as sources of a teaching in terms of being experts or upholders of the suttas and the Vinaya, to which the Tibetan version adds the Matrix- experts.220 A Dīrgha Āgama parallel in Chinese translation describes these elders as being learned and also experts in the suttas, the Vinaya and the Pātimokkha.221 In an Ekottarika Āgama parallel in Chinese, the learned elders are described as being capable in reciting the suttas and remembering the Dharma.222 Apparently, Chinese parallels to the Mahā,parinibbāna Sutta do not mention āgatâgama. This does not mean that the term is only found in the Pali suttas. In fact, two discourses in the Madhyama Āgama mention āgatâgama amongst the qualities of a learned monk. The Pali suttas, however, do not have any parallels to these two cases.223

5.2.1.4 In the Kusi,nārā Sutta (A 10.44), a monk who has not memorized the suttas and penetrated the teaching is exhorted to “(train to) master attainment” (āgamaṁ pariyāpuṇassu).224 Another Madhya- ma Āgama discourse similarly speaks of being capable at āgama recitation, which here is part of the de-

216 M 33/1:221,21 (= A 11.18/5:349,16) describes the elders as “learned, mastered the tradition, Dharma experts, Vinaya experts, Matrix experts,” bahu-s,sutā āgatâgamā dhamma,dharā vinaya,dharā mātikā,dharā (SD 52.6). SĀ 1249 (T2.343a25) and a sutta quotation in T1509 (T25.74a28) use the expression 多聞 duō wén; an individually tr discourse, T123 (T2.546c4), instead employs the expression 學問 xué wèn to qualify the elders as “learned.” 217 See prec n. 218 D 16/2:125,6 (SD 9) = A 4.180/2:169,18 (SD 3.1). 219 This statement is clearly in reference to Purāṇa, who visits Rājagaha after the 1st council, but when asked to accept its resolutions, gives this reply (Cv 11.1.11 @ V 2:288 f): see 3.11 n. It is possible that the 4 great references were formulated on account of Purāṇa here and inserted into the sutta. Mahā Kamma,vibhaṅga S (M 136) relates how the wanderer Potali,putta falsely claims to have heard the teaching directly from the Buddha himself, and presents wrong views which the novice monk (navaka,bhikkhu) Samiddhi is unable to put right (M 136,2/3:207). See S Dutt, Early Buddhist Monachism, 1984:18 f. 220 Fragment 360 fol 181 V6, Waldschmidt 1950:24: sūtradharā vina[yadharā]; and fol 182 R2 [vi]nayadh[arā]; Waldschmidt 1951:243 (§24.16): mdo sde ’dzin pa ’dul ba ’dzin pa ma lta bu ’dzin pa (transliteration standardized). 221 DĀ 2 (T1.17c15): 多聞 duō wén (T7 @ T1.195c23 and 196a3 has comparable refs to “learned”) and DĀ 2 (T1.17c27 + 18a11): 持法，持律，持律儀者 chí fǎ, chí lǜ, chí lǜ yí zhě. 222 EĀ 28.5 (T2.652b17 + c3 + c10): 誦經，持法…博學多聞 sòng jīng, chí fǎ ... bóxué duō wén. 223 MĀ 1 (T1.421b19, b21 + b23): 阿含及所得 ā hán jí suǒ dé and MĀ 95 (T1.577b8, b11 + b14): 阿含及其所得 ā hán jí qí suǒ dé; these refs to āgatâgama have no parallels in Dhammaññū S (A 7.64/4:114,3), SD 30.10, or Ṭhiti S (A 10.53.5:96,6). 224 A 10.44/5:80,23; a recommendation also given in V 2:249,16.

46 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists scription of a learned monk who may pride himself on his learning the texts.225 Understandably, by the time of the Chinese translations, āgama has evolved to mean “text” in the sense of scripture. Hence, it must be said that the expression āgama (singular) serves as one of the several terms to express that a learned monk was familiar with the teachings that were orally transmitted by the Buddha and his immediate disciples [5.2.1.1]. Analayo notes that, although this type of usage is attested in the Pali suttas and in parallel translations, they do not, as a rule, support it, meaning that the term is missing from these parallel texts.226 However, as noted [5.2.1.4], its old sense is still noticeable, albeit not in an apparent way, in the Madhyama Āgama.

5.2.1.5 Although the term āgatâgama fell out of use later, we still see the term āgama retaining its sense of being well versed in Dharma theory and practice even in Asoka’s time (c 268-232 BCE). The compound was shortened to āgama, but it retained the sense of attainment or tradition in keeping with what the Buddha taught. We can see the word in Asoka’s Rock Edict XII, which combines a reference to one being “learned,” bahuśrutā, with being “excellent (in the teaching),” kallāṇāgamā (Girnār version).227 Similarly, the Jain Vavahāra refers to a learned monk as “capable,” babbhāgamaṁ.228 The reason is quite clear why most of the parallel texts omit mention of āgatâgama or do not use it in the same way as the suttas. The term belongs to the early strata of Buddhism when adepts, arhats, knew the Dharma by heart, and readily recited or taught it. By the time of parallel translations in China, the early sense of the term was either forgotten or had taken on a different sense. In fact, in the Chinese translations, the term āgama lost its old sense as used in the suttas [see above here] and took on the sense of authorized or authentic scriptural texts: from a living oral tradition, it became a written scripture, even an authority in print. This is the development that we will explore next.

5.2.2 Āgama as textual authenticity

5.2.2.1 Before we examine the evolution of āgama in the Chinese translations a few centuries after Asoka’s time, there is an important evolutionary stage of the term which we need to examine. This will help us understand how authentication by realization in early Buddhism evolved into authority by testa- ment, or authorization of the printed word in Chinese Buddhism. The phrase āgatâgama, “one who has mastered the attainments,” is resolved as āgata + āgama; both of which are derived from the root √GAM OR √GACH, to go. Āgata, “(has) come, arrived, reached, accomplished” is the past participle of gacchati,”to go”; and āgama has at least 2 senses: (1) coming, arrival, access, attainment; (2) tradition, sacred knowledge, scriptural learning. (1) is the older sense, which refers to the “attainment” of the path (streamwinning, etc) or the “realization” of the path, that is, arhathood; while (2) is the later sense of “textual tradition.”229

5.2.2.2 In the suttas, the term āgatâgama refers to the authority of realization (paṭivedha): these are teachings reflecting the Buddha’s awakening that is reflected, re-enacted, in the awakening of the

225 MĀ 85 (T1.561b27, b28, c1, c2): 諳阿含 ān ā hán. Its parallel at M 113/3:39,18+31 has two separate cases, where a monk may be “deeply learned” (bah-s,suta) or be a “Vinaya-expert” (vinaya,dhara), without any ref to āgama. 226 Analayo 2016:12. 227 Bloch 1950:123,29, Girnār: bahusrutā ca assu kallāṇāgamā ca; Kālsī: bahuṣṣuta cā kayyānāgā ca; Ṣāhbāzgar- hī: bahuśruta ca kalaṇagama ca; Mānsehrā: bahuśruta ca kayaṇagama ca. 228 Vavahāra 1.35: Schubring 1918:15,4+6; cf Caillat 1965:50. I take Amg babbha as cognate with P bhabba. 229 See Also Analayo 2015a:50. http://dharmafarer.org 47 SD 58.1 The Oral Tradition of the Early Buddhists arhat. Both these awakenings are the same; the only difference is that the Buddha is the first to arise in the world, and the arhats follow after—as clearly stated in the Sambuddha Sutta (S 22.58).230

5.2.2.3 In the suttas, āgatâgama literally means “(one who has) come to the mastery, that is, the realization of the Dharma,” a broad term referring to an understanding of the teaching that brings us to the path up to arhathood. Hence, it refers to those who have such an understanding as those who have “mastered the tradition,” that is, the “attainment” (agama) of the path from streamwinning to arhathood. The word “tradition” here is synonymous with “lineage” (vaṁsa) in the early Buddhist sense of ariya,- vaṁsa, “the lineage of the noble ones.” This is a sutta term for the true practice of renunciation, that is, through the following ways, as stated in the Ariya,vaṁsa Sutta (A 4.27):231

(1) contentment regarding robes; cīvara,santosa (2) contentment regarding almsfood; piṇḍapāta,santosa (3) contentment regarding lodging; senâsana,santosa (4) delighting in the cultivation of good and the abandonment of bad. bhāvanā,pahān’āramatā

A disciple (monastic or lay) is said to be āgatâgama when he has attained the path of awakening. If he is a monastic (a monk or a nun), he is also said to be of “the noble lineage” (ariya,vaṁsa), in the sense of living a truly renounced life, wisely dependent on the 4 supports (catu paccaya)—robes, almsfood, lodging and health support.232 A lay disciple who is āgatâgama lives a simple life of moral virtue and mindfulness which are the bases for his great wisdom of the Buddha-word.

5.2.3 The 4 great references for the Dharma

5.2.3.1 An important textual evidence for the existence of the canon based on realization or attainment (āgatâgama) is the 4 “great references” (cattāro mahā’padesa), the criteria for establishing the authenticity of an early Buddhist text, such as that described in the Mahā,parinibbāna Sutta (D 16).233 The compound mahā’padesa means great “source” or “authority” (apadesa),234 and refers to the sources from whom a teaching may be accepted as Buddha-word. These 4 references or authorities for the authenticity of the teaching are:

(1) the Buddha himself; (2) a sangha (monastic community) (along with its experts and elders) dwelling in a certain monastery;235 (3) certain learned elder monks dwelling in a certain monastery;236 or (4) a single learned elder monk dwelling in a certain monastery. 237

230 S 22.58/3:65 f (SD 49.10). 231 A 4.27/2:27; D 33,1.11(9)/3:224; Nm 2:497; Nc 107; Pm 1:84. 232 See Sabb’āsava S (M 2,13-16), SD 30.3; Santuṭṭhi S (A 4.27), SD 104.8. 233 Mahā,parinibbāna S (D 16,4.7-4.11/2:123,30-126,5), SD 9, = Mahā’padesa S (A 4.180/2:167,31-170,19), SD 31(2.2). 234 DP sv. 235 Amukasmīm nāma āvāse saṅgho viharati sa-t,thero sa,pāmokkho. It is likely that pamokkha here are not only “monastic head” but are also Vinaya expert, Wynne (2004:100 n11) mentions the expression vinaye pāmokkho (which he refers to Gombrich 1992). However, I am unable to locate in CSCD. However, there is the phrase vināya- ko aggo pamukho pamokkho, in connection with the fire-worshipping Kassapa brothers (V 1:24 passim, 32) where it has the sense of “leader” in a string of (near)synonyms. 236 Amukasmiṁ nāma āvāse sambahulā therā bhikkhū viharanti bahu-s,sutā āgatâgamā dhamma,dharā vinaya,- dharā mātikā,dharā. (D 2:125,5)

48 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

Although there are 4 possible sources from whom a teaching may be properly accepted, in each case the method of establishing the authenticity of the teaching in question is the same, as laid out in this passage where a teaching is said to have been received directly from the Buddha, thus:238 4.8 (1) “Here, bhikshus, a monk might say this: ‘Avuso, I heard and received this teaching from the Blessed One’s own mouth: “This is the Dharma, this is the discipline, this is the Teacher’s teaching.”‘239 Then, bhikshus, you should neither approve nor disapprove of his word. Neither approving nor disapproving, his word and expression should be carefully studied and checked against the Sutta [teaching] and examined (for conformity) against the Vinaya [discipline]. (2) If they, on such checking and examining, are found to conform neither to the Sutta nor to the Vinaya, then it should be concluded: ‘Surely, this is not the Blessed One’s word. It has been wrongly understood by this monk,’ and the matter is to be rejected. But where, on such checking and examining, they are found to conform to the Sutta and the Vinaya, it should be concluded, “Surely, this is the Blessed One’s word. It has been rightly understood by this monk. Bhikshus, remember this as the 1st great reference. (D 16,4.8), SD 9240

5.2.3.2 The above passage on the 1st great reference [5.2.3.1], notes Wynne (2004:101), does not state what is to be done with the dhamma or vinaya which is accepted as the Buddha-word (bhagavato vacanaṁ). But because it is stated that the rejected teachings are to be abandoned, we can suppose that the opposite was to be done with what had been accepted as Buddha-word, that is, if it is thought to agree with what has already been collected under the heading of “sutta” and “Vinaya,” it is to be added to them.

237 Amukasmiṁ nāma āvāse eko thero bhikkhu viharati bahu-s,suto āgatâgamo dhamma,dharo vinaya,dharo mātikā,dharo. (D 2:125,24) 238 Idha bhikkhave bhikkhu evaṁ vadeyya: “sammukhā m’etaṁ āvuso bhagavato sutaṁ sammukhā paṭiggahi- taṁ, ayaṁ dhammo ayaṁ vinayo idaṁ satthu,sāsanan’ti. Tassa bhikkhave bhikkhuno bhāsitaṁ n’eva abhinanditab- baṁ na-p,paṭikkositabbaṁ. Anabhinanditvā appaṭikkositvā tāni pada,vyañjanāni sādhukaṁ uggahetvā sutte osāre- tabbāni, vinaye sandassetabbāni. Tāni ce sutte osāriyamānāni vinaye sandassiyamānāni na c’eva sutte osaranti, na ca vinaye sandissanti, niṭṭham ettha gantabbaṁ: “addhā idaṁ na c’eva tassa bhagavato vacanaṁ; imassa ca bhikkhuno duggahitan’ti. Iti h’etaṁ bhikkhave chaḍḍeyyātha. Tāni ce sutte osāriyamānāni vinaye sandassiyamānāni sutte ceva osaranti, vinaye ca sandissanti, niṭṭham ettha gantabbaṁ: “addhā idaṁ tassa bhagavato vacanaṁ; imassa ca bhikkhuno suggahitan’ti. Idaṁ bhikkhave paṭhamaṁ mahā’padesaṁ dhāreyyātha. 239 This statement is clearly in ref to Purāṇa, who visits Rājagaha after 1st council, but when asked to accept its resolutions gives this reply (Cv 11.1.11 @ V 2:288 f): D 16,3.11 n (SD 9). It is possible that the 4 great references were formulated on account of Purāṇa here and inserted into D 16. Mahā Kamma,vibhaṅga S (M 136) relates how the wanderer Potali,putta falsely claims to have heard a teaching directly from the Buddha himself, and presents wrong views which the novice monk (navaka,bhikkhu) Samiddhi is unable to put right (M 136,2/3:207). See S Dutt, Early Buddhist Monachism, 1984:18 f. 240 D 16/2:124,3-19 (SD 9). The same 4 “great references” are found in Skt fragments of Mahāparinirvāṇa Sūtra, in almost identical words as in the Pali Sutta, although there is no Skt version of the mahā’padesa (Waldschmidt 1950:238-252). According to Frauwallner, Buddhism spread to the far NW-west of India through Asoka’s missions: “The mission of Kassapagotta, Majjhima and Dundubhissara gave origin to the Haimavata and Kaśyapīya. The mission of Majjhantika led to the rise of the Sarvastivadin. The Dharmaguptaka school is perhaps issued from the mission of Yonaka-Dhammarakkhita … And the community of Ceylon owes its origin to the mission of Mahinda.” (Frauwallner 1956:22 f). If this is so, then, the coincidence of the Mahāparinibbāna S and Mahāparinirvāṇa S implies that this method of establishing canonicity arose before 250 BCE. (Wynne 2004:101 n15) http://dharmafarer.org 49 SD 58.1 The Oral Tradition of the Early Buddhists

According to Cousins (2004), the passage shows that there are different methods for collecting sutta and Vinaya material. He interprets the passage as follows:

If something does not match with the vinaya (vinaye sandissanti), it should be rejected. This suggests an established and relatively defined set of vinaya rules such as we know to have existed from the comparative study of surviving vinaya works of various schools. Similarly something should be rejected if it does not enter into sutta (sutte otaranti). This is an unusual expression; it is best interpreted in the light of the Peṭakopadesa tradition where otaraṇā is one of the sixteen hara-s. It may there be taken as a particular method of exegesis which links a given discourse into the teaching as a whole by means of one of the general categories of teaching. The Peṭakopa- desa in fact specifies six possibilities: aggregates, elements, spheres, faculties, truths, dependent origination. Any one of these can be used to analyse the content of a discourse and their use will automatically place it in its context in the teaching as a whole. What is envisaged for sutta is not then a set body of literature, but rather a traditional pattern of teaching.241 Authenticity lies not in historical truth although this is not doubted, but rather in whether something can accord with the essential structure of dhamma as a whole.242

5.2.4 The 6 modes of conveyances (otaraṇa hara)

5.2.4.1 Wynne discusses Cousins’ explanation of the great reference [5.2.3.1] which we should carefully consider in this section.243 For Cousins, then (notes Wynne), the difference between the verbs sandissati (in the phrase vinaye sandissanti) and otarati (in the phrase sutte otaranti) is that sandissati means “match” and implies that the “Vinaya” with which some new teaching is to be matched is relatively fixed, whereas otarati means “enter into” and implies that the “sutta” with which a new teaching is to be compared is “not a set body of literature, but rather a traditional pattern of teaching” [which may be said to be āgama in the early Buddhist sense] [5.2.1]. Therefore, Cousins implies that doctrinal coherence rather than historical truth was the motivating factor of those who put together the collection of doctrinal discourses called “sutta.” Is this an accurate estimation of this passage? asks Wynne. The difference between the verbs used to describe the act of comparing teachings with either “sutta” or “Vinaya” is certainly of some significance. Cousins’ suggestion that otarati ought to be interpreted in the light of the Peṭakopadesa (Peṭk) definition of otaraṇā makes good sense.244 It probably means, as Cousins indicates, that the doctrinal content of a new teaching under consider- ation is to be compared with the doctrinal content of a body of oral literature called “sutta,” in one of the 6 categories of descent (into the path),245 which are as follows:246

(1) the aggregates or categories (khandha) the 3 trainings (in moral virtue, concentration, wisdom); (2) the elements (dhātu) the formations (saṅkhāra) that is the “mind-element” (dhamma,dhātu), free from defilements and from being; (3) the sense-bases (āyatana); the bases are free from influxes and being;

241 This is what we may consider as āgama [5.2.1]. 242 Cousins, “Pāli oral literature,” 1983:2 f. 243 Wynne 2004:102-104. 244 Peṭk §§366-431/98-110 (Peṭk:Ñ 132-154); Nett §3.12/63-70 (Nett:Ñ 93 f). 245 Ñāṇamoli in his trs of Peṭk and Nett render otaraṇo haro as “modes of entry.” 246 Peṭk §366/98; Nett §3.12/351/63.

50 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

(4) the spiritual faculties (indriya); the learner’s faculties, that is, the eradication of the self- identity view (attainment of streamwinning); (5) the truths (sacca); the 4 noble truths; (6) (the 2 modes of) dependent arising the learner’s 5 faculties are spiritual knowledge (vijjā) that (paṭicca,samuppāda). ends ignorance and the rest of the links of dependent arising.

The Petakopadesa says: “There is no sutta, whether verse or prose, that does not show one or other other 6 ideas. For the whole extent of the teaching is about either the aggregates or the bases or the faculties or the truths or dependent arising.”247 In fact, both the Peṭakopadesa and the Netti-p,pakaraṇa explain elaborately how these ideas or teachings (dhamma) interlinked and embody one another. These teachings reflect an early period when the various ideas or teachings have not been categorized as dis- crete categories as in the nikāyas [6].

5.2.4.2 This shows that the collection we now know as sutta was still not a “set body of literature.” The 4 great references were current at this time, that is, to the sutta or teaching as a whole. However, the 4 great references still do not inform us how the suttas, individually or as a collection, as we know them, were composed or transmitted. A careful examination of the wording of the passage suggests that the works comprising “sutta” were transmitted word for word. We can deduce this because we are told that the “words and letters” (pada,- vyañjanāni) of the teaching here were to be “learnt correctly” (sādhukaṁ uggahetvā) before we decide its authenticity. Wynne reminds us that if attention was to be paid to the words and letters of these teachings, it implies that the content of what was known as “sutta” was also transmitted by paying a similar attention to its words and letters, that is, it was transmitted word for word. The passage therefore shows that the accuracy with which a body of literature called sutta was meant to be transmitted was very high, down to the letter. At this stage, the suttas, then, were not a fixed body of literature; for, it could be supplemented by comparing its already established doctrinal content with the doctrinal content of new teachings, which could then be added to it. Clearly, the early Buddhists at least persevered to transmit the teachings accurately.248

5.2.4.3 However, notes Wynne, contrary to what Cousins thinks, the verb sandissati, “to conform,” cannot mean that the set of Vinaya rules was relatively fixed. Instead, it seems that sandissati here is simply the usual verb used to state that a person conforms to certain ethical or religious practices,249 or that certain practices or states are found “in” a person or persons.250 [5.2.4.1]

247 N’atthi taṁ suttaṁ vā gāthā vā vyākaraṇaṁ vā imesaṁ channaṁ dhammānaṁ aññatarasmiṁ na saṁdissati. Ettāvatā esā sabbā desanā yā tā khandhā vā dhātuyo vā āyatanāni vā indriyāni vā saccāni vā paṭicca,samuppādo vā. (Peṭk §366/98) 248 Wynne 2004:102 f. 249 Eg: “Now, … do you live with your teacher in conformity with this peerless accomplishment of knowledge and conduct?” (api nu tvaṁ imāya anuttarāya vijjā,caraṇa,sampadāya sandissasi sâcariyako’ti?) D 3,2.4/1:102,10 (SD 21.3); “Brahmin, so long as these 7 conditions for non-decline endure amongst the Vajjīs, so long as the Vajjīs conform to these 7 conditions for non-decline … ” (yāvakivañ ca brāhmaṇa ime satta aparihāniya dhammā vajjīsu ṭhassanti, imesu ca sattasu aparihāniyesu dhammesu vajjī sandissanti), D 16,1.5/2:75,27-29 (SD 9) = A 7.21/4:17,8+20,20 (SD 55.11), A 7.23/4:22,7+ 24 (SD 106.2), A 7.24/23,9 (SD 106.3); http://dharmafarer.org 51 SD 58.1 The Oral Tradition of the Early Buddhists

It is understandable, therefore, that in the passage in question, it is asked if the words and letters of the teaching “conform” (sandissanti) to the “Vinaya,” for this is the verb that was to be used when considering a thing’s conformity to religious practices. There is no implication that the Vinaya was fixed.

5.2.5 The 4 great references for the Vinaya

5.2.5.1 Although Cousins argues that the passage on the 4 references shows variability in the fixing of suttas but not the contents of the Vinaya, there is, in fact, no such implication. Instead, we are told that the early Buddhist literature consisted of primitive collections called sutta and vinaya, and we can only deduce that both of these were periodically expanded by the addition of new material. This is what is called dhamma,vinaya, “the teaching and the discipline,” in the suttas. In order for new material to be accepted into these collections, they were learnt, that is, memorized, word for word, and then compared with the content of the existing collections. If the comparison showed that the new agreed with the old, it was added to it, and no doubt transmitted word for word. If the passage on the 4 references reflects the actual practice of early monastic Buddhism, it is hard to imagine that improvisational methods of oral transmission could ever have been used [4.2.1], for such methods do not guarantee the accuracy to the letter demanded by standard of the 4 references.251

5.2.5.2 What has been said about the suttas above applies just the same to the way in which the Vinaya was formed. It is clearly not an “established and relatively well defined set of vinaya rules,” as suggested by Cousins, because, like the suttas, the Vinaya, too, was, in the Buddha’s time not a fixed code of rules, case-histories and decisions, we know today as the “Vinaya.” Interestingly, in all the scholarly excursus on the great references (mahā’padesa), they were applied only to the Dharma. The Vinaya, too, has its own set of 4 great references. The Vinaya Mahā,vagga, chapter 6, on medi- cines, records monks had scruples “what is permitted by the Blessed One, what is not permitted” (kiṁ nu kho bhagavata anuññātaṁ kiṁ ananuññātaṁ).252 In fact, the 4 great references for the Vinaya, considering its legal nature, are more comprehensive and exacting, laid out as follows:

(1) yaṁ bhikkhave mayā idaṁ na kappatîti Whatever, bhikshus, has not been disallowed appaṭikkkhittaṁ, tañ ce akappiyaṁ anulometi by me, thus: “This is not allowable,” if it fits in kappiyaṁ paṭibāhati, taṁ vo na kappati. with what is not allowable, if it goes against the allowable, that is not allowable to you. (2) yaṁ bhikkhave mayā idaṁ na kappatîti Whatever, bhikshus, has not been disallowed

“These states are found in me, and I am engaging in them” (saṁvijjante te ca dhammā mayi ahañ ca tesu dhammesu sandissāmîti), M 129,3.2/3:163,23 etc (SD 2.22) = S 47.29/5:177,19, 55.3/5:345,17+29 (SD 23.16), 55.39/5:- 397,7, 55.53/5:407,28 (SD 46.4); “Bhante, these manifestations of faith in one who has faith that the Blessed One has spoken of, exist in this monk, and this monk conforms to them” (yan’imāni bhante bhagavatā saddhassa saddhā,padānāni bhāsitāni, saṁvijjanti tāni imassa bhikkhuno, ayañ ca bhikkhu etesu sandissati), A 11.14/5:340,31 (SD 99.2). 250 Eg: “Master Gotama, do the brahmins today observe the Brahminical traditions of the ancient brahmins?” (sandissanti nu kho bho gotama etarahi brāhmaṇā porāṇaṁ brāhmāṇaṁ brāhmaṇa,dhamme’ti?), Sn 2.7/50,18; “Thus, … such states that are unwholesome … are found among some kshatriyas … brahmins, too … vaishyas, too … found amongst some shudras, too” (ye’me dhammā akusalā … khattiye pi te idh’ekacce sandissanti … [brāhmaṇe pi … vesse pi…] sudde pi te idh’ ekacce sandissanti), D 27,5/3:82,11-20 (SD 2.19); “Bhikshus, these 5 ancient practices are today seen amongst dogs but not among brahmins” (pañc’ime bhikkhave porāṇā brāhmaṇa,dhammā etarahi sunakhesu sandissanti no brāhmaṇesu), A 5.191/3:221,12 f. 251 Wynne 2004:103 f. 252 Mv 6.40 (V 1:250 f).

52 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

appaṭikkkhittaṁ, tañ ce kappiyaṁ anulometi by me, thus: “This is not allowable,” if it fits in akappiyaṁ paṭibāhati, taṁ vo kappati. with what is allowable, if it goes against the unallowable, that is allowable to you. (3) yañ ca bhikkhave mayā idaṁ kappatîti And whatever, bhikshus, has not been permit- ananuññātaṁ, tañ ce akappiyaṁ anulometi ted by me, thus: “This is allowable,” if it fits in kappiyaṁ paṭibāhati, taṁ vo na kappati. with what is not allowable, if it goes against the allowable, that is not allowable to you.

(4) yaṁ bhikkhave mayā idaṁ kappatîti Whatever, bhikshus, has not been permitted ananuññātaṁ, tañ ce kappiyaṁ anulometi by me, thus: “This is allowable,” if it fits in akappiyaṁ paṭibāhati, taṁ vo kappati. with what is allowable, if it goes against the unallowable, that is allowable to you. (Mv 6.40.1 @ V 1:250,34-251,6)

The great references for the Vinaya—like the great references for the Dharma—attests to the fact that the Vinaya, too, was not, during the Buddha’s time, perhaps up to Asoka’s time, a fixed code of laws, cases and decisions. New rules were made, old ones modified, even abrogated, with new case histories and decisions made and recorded—that is, until the canon was closed by Asoka’s time.253

5.2.6 A list of lists

5.2.6.1 Probably late in the Buddha’s life, the term āgama shedded its earlier meaning of “attainment or realization,” and evolved into what we today understand as “scripture.” This interesting evolutionary stage of the early Buddhist texts was highlighted by the categories of 3, 4, 9 or even 12 limbs (aṅga) [5.4], which was, in turn, superseded by the Tipiṭaka (“3 baskets”) scheme of arranging the teachings into the “baskets” of the Vinaya (vinaya,piṭaka), suttas (sutta,piṭaka ) and the Abhidhamma (abhidhamma,piṭaka) [5.5.2]. Let us now examine these developments, leading to the “limbs” (aṅga) of the teaching.

5.2.6.2 When we first come into contact with Buddhist teachings, we are likely to hear about a few basic Buddhist lists: the 3 jewels (ti,ratana), the 4 noble truths (catu ariya,sacca), the 5 precepts (pañca,- sīla) or the noble eightfold path (ariya aṭṭh’aṅgika magga). As we continue to listen to Buddhism or read about it, we notice that there are more lists: the 4 paths (catu magga), the 4 dhyanas (catu jhāna), the 5 aggregates (pañca-k,khandha), the noble individuals (aṭṭha ariya,puggala), the 10 unwholesome courses of karma (dasa akusala kamma,patha), the 10 fetters (dasa saṁyojana), and so on.254

5.2.6.3 Rupert Gethin, in his paper “The Mātikas” (1992)255 discusses how the making of lists in early Buddhism served as a “mnemonic technique.” Practically, most of the suttas found in the 5 Nikayas may be regarded as teachings based on some kind of lists, and that many suttas can be summed up by listing the teachings they give.

253 Asa a text, the Vinaya may be closed; however, the sangha continues to function as a “legal person,” such as performing ordination, adjudicating cases of Vinaya offences, conducting sangha acts (such as establishing “bor- ders” (sīmā) for the consecrated convocation hall), and so on. 254 There are a number of Buddhist dictionaries that give lists of lists, eg, Prayudh Payutto, Dictionary of Buddh- ism, Bangkok, 1972 (many reprs), pp1-359; Buswell & Lopez (eds), The Princeton Dictionary of Buddhism, 2014: 1065-1102 List of lists; Piya Tan, Sutta Discovery Numerical Index, 2010- . 255 R M L Gethin, “The Mātikas: Memorization, mindfulness, and the list,” in (ed) The Mirror of Memory: Re- flections on Mindfulness and Remembrance in Indian and Tibetan Buddhism, Albany, NY: SUNY, 1992b: 149-172. http://dharmafarer.org 53 SD 58.1 The Oral Tradition of the Early Buddhists

Two such lists are often presented as introductory accounts of Buddhism, even as its foundations, are the noble eightfold path and the 4 noble truths. This is understandable since these 2 teachings are the basis for the Buddha’s 1st discourse, the Dhamma,cakka Pavattana Sutta (S 56.11).256 In fact, the Buddha once declares that he only teaches “suffering and its ending” [1.2.2.2].

5.2.6.4 “But why are there so many other lists in Buddhist thought and literature?” asks Gethin (1992:149). Various scholars have highlighted the usefulness of these lists as mnemonic devices,257 and it seems clear that the numerous lists in early Buddhist literature have to do with its being an “oral literature.” The Buddha gave only verbal teachings like the teachers of his day. These teachings were then collected, edited, disseminated and memorized. This process of textual composition is somewhat like how we today would transcribe talks by a teacher, edit it so that it is suitable to appear in print and disseminated. An important difference is that the early Buddhist process of textual composition was done orally throughout, and only subsequently became fixed in the form of written, and today, printed, texts.

5.2.6.5 In an interesting way, these early Buddhist texts redefined or extended what we understand by the term, literature—a “literature” that was composed orally and only subsequently became fixed in the form of written texts. Etymologically, the word literature comes from Latin, littera, “letters,” which make it literally to refer to “something written (with letters).” The online Merriam-Webster dictionary defines “literature” as follows: “writings having excellence of form or expression and expressing ideas of permanent or universal interest.” On account of the early Buddhist texts, we need to broaden this definition to: “textual records having excellence of form or expression, and expressing ideas of enduring or universal interest.” This definition will then do justice to ancient Indian culture (indeed, any culture) which, as a whole, was oral by nature.258 Ancient Indian learning was oral by nature, a direct communicating between teacher and pupil. This early system had “a penchant for analyzing something in terms of a neatly categorized list … characteristic of much of traditional Indian learning, and the oral origins of Indian learning continued to inform its structure long after its exponents had begun to commit it to writing.”259 Gethin reflects thus:

One only has to reflect for a minute on the difficulties of composing a talk or a discourse without the aid of pen and paper, or without access to computers and word processors, to begin to appreciate what a convenient solution the list is. A list immediately imparts to the discourse a structure that makes it more easily remembered by the one giving the talk. At the same time a talk based on lists is easier to follow and remember for those listening. With a list one has a certain safeguard against losing one’s way in a talk or forgetting sections of it. Thus if I go to a talk by the Buddha on the noble eightfold path and later find I can only remember five of the eight “limbs,” then, provided that I remember that buddhas always talk about eightfold paths, I will at least know that I have forgotten something and do not remember the talk in full. (Gethin 1994b:149 f)

5.2.6.6 Although lists might be a feature of ancient Indian literatures in general, it is probably no exaggeration to say that the early Buddhists were the most prodigious makers of lists. It would be inter-

256 S 56.11/5:420-424 = Mv 1.6.17-31 (V 1:10-12), SD 1.1. 257 S Collins, Selfless Persons, 1982:109; L S Cousins, “Pali Oral Literature,”1983:1-11 (esp 3 f). 258 See, eg, E Frauwallner. History of Indian Philosophy, 2 vols, 1973, 1:20-23. 259 Gethin 1992b:149.

54 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists esting to study how the early Buddhist texts formed themselves around lists, and how these lists proliferate and interconnect. An obvious starting point is the list of the 4 noble truths. The bare statement of this list is as follows:

[The 4 noble truths:] This is the noble truth that is suffering. This is the noble truth that is the arising of suffering, This is the noble truth that is the ending of suffering, This is the noble truth that is the path leading to the ending of suffering.260 This terse statement on the 4 noble truths is often elaborated in the suttas as follows:

(1) Now this, bhikshus, is the noble truth [reality] that is suffering: birth is suffering; decay is suffering; disease is suffering; death is suffering; [grief, lamentation, physical pain, mental pain and despair are suffering];261 to be with the unpleasant is suffering; to be without the pleasant is suffering; not to get what one desires is suffering, in short, the 5 aggregates of clinging262 are suffering.

(2) Now this, bhikshus, is the noble truth [reality] that is the arising of suffering: it is this craving that leads to renewed existence [rebirth], accompanied by pleasure and lust, seeking pleasure here and there; that is to say, craving for sensual pleasures, kāma,taṇhā craving for existence, bhava,taṇhā craving for non-existence [for extinction]. vibhava,taṇhā

(3) Now this, bhikshus, is the noble truth [reality] that is the ending of suffering: it is the utter fading away and ending of that very craving, giving it up, letting it go, being free from it, being detached from it.

(4) Now this, bhikshus, is the noble truth [reality] that is the path leading to the ending of suffering: it is this very noble eightfold path, that is to say, right view, right thought, right speech, right action, right livelihood, right effort, right mindfulness, right concentration.263

260 See Saṅkāsana S (S 56.19/5:430), SD 53.32; Pm 1:37. Cf D 33,1.11(12)/3:227 the 4 knowledges (ñāṇa); 34,1.5- (9)/3:277 the 4 things to be realized (cattāro dhammā sacchikātabbā). For the various ways in which the truths are listed in the suttas and problems in tr them, see K R Norman, “The four noble truths: A problem of Pali syntax,” 1982:377-391. 261 Found in Ee Se (esp in daily chant) and other MSS but not in Be and Ce. This addition is prob late, as it is not found in the Chin versions. The Chin often speak of the “8 sufferings” (bākǔ), as listed above. 262 Pañc’upadāna-k,khandha, viz, form, feeling, perception, formations and consciousness (S 3:47; Vbh 1). What is briefly mentioned here is elaborated in the 2nd discourse, Anatta,lakkhaṇa S (S 22.59), SD 1.2. 263 Dhamma,cakka Pavattana S (S 56.11,5-8/4:421 f), SD 1.1. http://dharmafarer.org 55 SD 58.1 The Oral Tradition of the Early Buddhists

5.2.6.7 We can now at once see that this definition of the 4 noble truths links into other sutta lists. Thus, the 1st truth is summed up as the list of the 5 aggregates of clinging (pañc’upādāna-k,khanda); the 2nd truth is explained in terms of various kinds of “thirst” (taṇhā), that is, the 3 kinds;264 the 3rd truth consists in the ending of these very same 3 kinds of thirsts (that is, nirvana); the 4th truth is famously summed up as the noble eightfold path.265 This first stage of analysis of the list of the 4 truths links into 3 further lists. Now that we know that suffering is embodied as the 5 aggregates of clinging, we would like to know what the Buddha teaches about them. From the suttas, we learn that they are listed as: form (rūpa), feeling (vedanā), perception (saññā), formations (saṅkhārā) and consciousness (viññāṇa).266 We see various definitions of these 5 aggregates, and these definitions, in turn, give us still more sutta lists. Thus, form or physical form is the 4 “great elements” or “great essentials” (mahā,bhūta), namely, the elements of earth, water, fire and wind.267 Feeling consists of the 3 feelings that are pleasant, painful, and neither-painful-nor-pleasant.268 Perception is described as arising in 6 ways, that is, through perceiving forms, sounds, smells, tastes, touches and thoughts. Perception is essentially how we recognize a sense-object and give our experience its hedonic tone.269 Then, there are 3 formations or volitions, namely, those of body, speech and mind. This refers to our mental process when we create karma through what we do, say or think.270 Consciousness is of 6 kinds, depending on the eye, ear, nose, tongue, body or mind. Simply, this is how we sense things at the most basic level.271

Alternatively, both feeling and formations (volitions) may be understood as being of 6 kinds each: feeling born of contact through the eye, ear, nose, tongue, body or mind and volition associated with shapes, sounds, smells, tastes, touches, or thoughts.272 These explanations by way of 6 classes based on another well known sutta list, that of the 6 sense- bases (saḷ-āyatana) or simply, the 6 senses.273 Appropriately enough, the 1st truth is occasionally summed up, not in terms of the 5 aggregates, but in terms of these 6 sense-bases.274

5.2.6.8 Like the 1st truth, the 4th truth also demands and receives considerable elaboration. The noble truth that is the path leading to the end of suffering is said to consist of 8 “limbs” (aṅga). Once

264 By the close of the Nikaya period, "thirst” (taṇhā) or craving, achieved the status of list in its own right, as “the 3 kinds of thirst (cravings),” those for sensual pleasure, for existence, for non-existence (tisso taṇhā: kāma,- taṇhā bhava,taṇhā vibhava,taṇha, D 33,1.10(16)/3:216). 265 The eightfold path itself branches out into at least 2 more important lists: those of the 3 trainings (sikkhā) (in moral virtue, in concentration and in wisdom), and the 4 paths (magga) of streamwinning, once-returning, non- returning and arhathood. 266 For refs and a fuller discussion related to the 5 aggregates, see R Gethin, “The five khandhas: Their treatment in the Nikāyas and early Abhidhamma,” 1986:35-53; also SD 17. 267 For teaching on the 4 elements, see Mahā Hatthi,padôpama S (M 28,5-38/1:185-191), SD 6.16. On form as aggregate, see SD 17.2a. 268 Cūḷa Vedalla S (M 44,22-24/1:302 f), SD 40a.9. On feeling (vedanā) as aggregate, see SD 17.3. 269 Upādāna Parivaṭṭa S (S 22.56,13-15/3:60), SD 3.7. On perception (saññā), see SD 17.4. 270 Cūḷa Vedalla S (M 44,13-15/1:301), SD 40a.9. On formations (saṅkhārā), see SD 17.6. 271 Upādāna Parivaṭṭa S (S 22.56,19-21/3:61), SD 3.7. On consciousness (viññāṇa), see SD 17.8a. 272 Upādāna Parivaṭṭa S (S 22.56,10-12/3:59 f), SD 3.7. 273 See esp Saḷāyatana Saṁyutta (S 35/4:1-203); also Saḷ,āyatana Vibhaṅga S (M 137) + SD 29.5 (1.2). 274 (Sacca) Ajjhattik’āyatana S (S 56.14/5:426), SD 53.27.

56 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists more these 8 limbs are explained in more detail in the suttas, and once more these explanations refer to yet more sutta lists.275 Thus, right view (sammā,diṭṭhi) is the right knowledge of suffering, its arising, its ending and the path to its ending—in other words, penetrating the 4 truths. In spirit, then, right view underpins every one of the other 7 limbs of the path, making them “right” (sammā).276 Right thought (sammā,saṅkappa) is explained as being of 3 kinds, namely, thoughts that are free from desire, free from hatred, and free from cruelty; they feature in the Saṅgīti Sutta (D 33) as “the 3 wholesome thoughts.”277 On a deeper level, right thought refers to the intention behind all our actions, mental, verbal and bodily; hence, sammā saṅkappa is also rendered as “right intention.”278 Right speech (sammā,vācā) is speech that abstains from the 4 wrong kinds of speech, that is, false speech (lying), divisive speech, unpleasant speech and frivolous chatter. These wrong kinds of speech are subsumed under the 4th of the 5 precepts that highlights the 1st wrong speech. [See following.] Right action (sammā,kammantā) is action that refrains from the 3 bodily misdeeds of killing living beings, taking the not-given, and sexual misconduct. These, in turn, for the first 3 of the 5 precepts, the natural morality that are karmically potent; hence, binding on all, whether monastic or lay.279 Right livelihood (sammā,ājīva) is explained simply as “abandoning wrong livelihood and making a living by means of right livelihood,” which basically means at least not breaking any of the 5 precepts (for both monastic and the laity). For monastics, right livelihood means not earning a living in any way, but fully living the life of renunciation in keeping the letter and spirit of the Vinaya and renunciation. Right effort (sammā,vāyāma) is explained by way of a stock sutta formula detailing 4 kinds of effort that are elsewhere called “the 4 right strivings” (cattāro samma-p,padhānā), that is, the efforts to refrain from doing bad not yet done, to abandon bad that has been done, to cultivate good not yet done, and maintain good already done.280 Right mindfulness (sammā,sati) is explained by another stock sutta formula concerning the 4 kinds of contemplation (anupassanā) that are usually called “the 4 focuses of mindfulness” (cattāro satipaṭ- ṭhāna), relating to body-based meditation, feeling-based meditation, mind-based meditation and reality-based meditation.281 Lastly, right concentration (sammā,samādhï) is defined as the successive attainment of the 4 dhyanas (jhāna). This is the stage when the practitioner, by his own mental effort, is able to rise above the limits of the body, attaining the pure mental level so that he enjoys a blissful life here and now, and with wisdom, he would be able to attain the path.

5.2.7 Meditation and the path pericope

5.2.7.1 Once we begin to know the Dharma, we start to shed our ignorance. What used to comfort us with false hope and hubris is now a profound hollowness that we want to urgently fill with true reality and its attendant freedom. We seek answers for questions we do not even know how to ask, but questions we have; doubts we have. As our understanding of the Dharma deepens, we begin to know the kind of questions to ask, when to ask them.

275 Mahā Satipaṭṭhāna S (D 22,21/2:311), SD 13.2; Sacca Vibhaṅga S (M 141/3:251), SD 11.11; (Magga) Vibhaṅga S (S 45.8/5:8-10), SD 14.5. 276 On right view, see SD 10.16 (1.8). 277 D 33,1.10(6)/3:215: tayo kusalā saṅkappā: nekkhamma,saṅkappo avyāpāda,saṅkappo avihimsā,saṅkappo. 278 On right intention, see SD 10.16 (2). 279 On the 5 precepts (pañca,sīla), see Saṅgīti S (D 33,2.1(9)/3:234); Dīgha,jānu S (A 8.54,13), SD 5.10; Veḷu,dvār- eyya S (S 55.7), SD 1.5 (2); Sīlânussati, SD 15.11 (2.2); SD 21.6 (1.2); SD 37.8 (2.2). 280 See in particular the Samma-p,padhāna Samyutta (S 49/5:244-248). On right effort, see SD 10.16 (6). 281 Classically, in Mahā Satipatthana S (D 22/2:290-315), SD 13.2, and Satipaṭṭhāna S (M 10/1:5-63), SD 13.3. http://dharmafarer.org 57 SD 58.1 The Oral Tradition of the Early Buddhists

Our quest for answers, for explanation and exposition, uncover more lists, and we link more lists together; the picture become bigger and clearer. But the missing parts as still very glaring: our quest then is for completeness, fulfilment. By now, we know better than the answer lies within our own mind. For this, there is no better tools than the focuses of mindfulness. A detailed and helpful exposition of the 4 focuses of mindfulness is found in the Satipaṭṭhaāna Sutta (M 10). The 1st focus of mindfulness, the “contemplation of body in the body” (kāye kāyânupassanā), consists of various meditations that are body-based. The best known of them is the mindfulness of breathing (which is itself expanded in another sutta).282 5.2.7.2 Continuing this pursuit of sutta lists, we briefly return to the 8th limb of the path, right concentration. We have already noted its stock passage as the 4 dhyanas (jhāna) [5.2.6.8]. This 4-dhyana pericope is part of a larger path pericope, that appears to be an abridgement (peyyāla)283 of a fuller passage that forms the focus of what is clearly the classic sutta account of the Buddhist path, from renunciation, through meditation (jhāna) to arhathood.284 In this context, the passage on the 4 dhyanas is at once prefaced by the abandoning of the 5 hindrances. In the Dīgha pericope, it is followed by the attainment of what are later known as the 6 super- knowledges (abhiññā); in the Majjhima pericope, by what are later known as the 3 knowledges (vijjā).285 Both sets refer to the mental powers and liberation (all called “knowledge”) gained by true meditation, the only difference is in their number. The last of these knowledges, in either case, (the knowledge that of arhathood) is the full understanding of the 4 noble truths: that of suffering, its arising, its ending, and the path to its ending: we thus return to the root teaching set. At this stage, however, we are transformed by all this understanding: we have eradicated mental suffering through being liberated from defilements in their most radical form,286 that of the 3 influxes (āsava), that is, those of sensual lust, existence and ignorance.287

282 The fourfold practice consists of (1) breathing in and out with a long breath, (2) breathing in and out with a short breath, (3) breathing in and out experiencing the whole body (of breath), and (4) breathing in and out stilling the breath. From viewpoint of the classical text, Ānâpāna,sati S (M 118,23-28/3:83-85), SD 7,13, and related suttas (Ānâpāna,sati) Ānanda S 1 + 2 (S 54.13+14/5:329-334), SD 95.12+13, this is only the first of what Majjhima Comy calls 4 “tetrads” (catukka); the treatment relates the 4 tetrads to the 4 focuses of mindfulness. 283 These abridgements (peyyāla) are not omissions, but they are so well known that they need not be laid out in full in the printed text. They are, in fact, recited in full each time, and hence we need to translate them in full, too, for the proper benefit of reading the suttas. On Cousins’ criticism of a commercial tr of the Nikāyas with such and other abridgements, see his “Review of the Middle Length Discourses of the Buddha, 1997; also SD 49.1 (6), SD 52.4 (1.3.3.3 f). 284 This 4-dhyana pericope occurs in full (though lost in the abridgements of the texts) 10 times of the 13 suttas of the 1st book of Dīgha (sīla,vagga): D 2,40-98/1:62-85 (SD 8.10), 3,2.2/1:100 (SD 21.3), 4,23/1:124 (SD 30.5), 5,27/- :147 (SD 22.8), 6,16/1:157 f (SD 53.4), 7,1/1:159 f (SD 53.41*), 8,28-30/1:171-174 (SD 73.12), 10,1.7-2.37/1:206-209 (SD 40a.13), 11,9-66/1:214 f (SD 1.7), 12,20-77/1:232 f) (SD 34.8). The pericopes in Poṭṭhapāda S (D 9), SD 7.14, and Tevijja S (D 13), SD 1.8, diverge after the passage on the 4th dhyana (jhāna). There is also a shorter Majjhima version of this path pericope: M 27,11-26/1:178-184 (SD 40a.5), 38,31-40/1:267-270 (SD 7.10), 51,12-27/1:344-348 (SD 32,9), 112,12-20/3:33-36 (SD 59.7), 125,13-29/3:134-137) (SD 46.3). 285 Saṅgīti S (D 33) lists the 3 vijjā (D 33,1.10(58)/3:220). Das’uttara S (D 34) lists the 3 vijjā (D 34,1.4(10)/3:275) as well as the 6 abhiññā (D 34,1.7(10)/3:282), stating, in either case, that they are “to be realized” (sacchikātabbā). See also PED svv abhiññā, vijjā. 286 This is a paraphrase of the nibbidā formula: longer version SD 20.1 esp (2.2.2); shorter version Alagaddûpama S (M 22,29), SD 3.13. 287 Saṅgīti S (D 33) lists 3 influxes (āsava): kām’āsava, bhav’āsava, avijjâsava (D 1.10(20)/3:216); also 4 influxes as “floods” (ogha): kām’ogha, bhav’ogha, avijj’ogha (D 33,1.11(31)) and as “yokes” (yoga): kāma,yoga, bhava, yoga, avijjā,yoga (D 33,1.11(31)); and the influxes extended into the 5 mental bondages (cetaso vinibandhā): (1) craving for sensuality, (2) attachment to one’s body, (3) craving for form, (4) gluttony and oversleeping, (5) living the holy

58 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

5.2.8 An early list of teachings (Gethin)288

5.2.8.1 The list of oral teachings that we have collected can be visually assembled in a tree diagram [Fig 5.2.8] to show how they are related. Familiar as this arrangement may be, it is not the only one there is. We can continue to explore more lists from any of these teachings. Indeed, even with a single list, we can build more lists with that list. It can all go back to the 4 noble truths. Teleologically, any of these lists is a portal and passage to the path of awakening. These are not lists for mere talk or study; they are what inspire us to go on our inner quest for awakening, or the light by

life for attaining heavenly rebirth: Ceto,khila S (M 16,8-12), SD 32.13; SD 51.10 (2.2). The latter 5 also overlap with the 5 mental hindrances (pañca nīveraṇa): Nīvaraṇa SD 32. Cf PED sv āsava. 288 This section, incl Fig 5.2.7, is based on R Gethin, “The Mātikās,” 1992b:153-158. http://dharmafarer.org 59 SD 58.1 The Oral Tradition of the Early Buddhists

60 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists which we begin that inward quest. Hence, each list, each step in a list, may turn back on itself, repeat itself, the part subsuming the whole. After all, it is in the very breath that we take in and give back that the Buddha finds awakening, that we will find awakening. All these lists are simply names and means we give to the natural breathwork that opens nirvana’s door. Hence, we are not immediately concerned with any chronological evolution of the sutta lists. What concerns us is how such lists and literature might have “appeared” to a monastic (how he would have heard or known it) towards the close of the sutta (or Nikāya) period. His perspective must have been syn- chronic: everything, as it were, is present, or can be teased out, starting with that text. Looking at the Dharma this way is just like looking at the Buddha—at a glance, we see awakening. Let the scholars toil with the evolution of these lists in time. As practitioners, we see even a single list as an opening (okāsa) out of the crowdedness (sambādha) of time and space, samsara.289

5.2.8.2 Having said that, we can now get down to mundane, time-bound, task of examining how these lists work. There are 2 basic ways in which the lists evolved and grew. Clearly, certain lists—such as those of the 4 truths, the eightfold path, the 5 aggregates of clinging and the 6 sense-bases—are more fundamental than others. They must each have evolved on their own in terms of actual personal practice. Only later, when they are taught as Dharma, they are shown to fit and work together, such as in the way that Gethin has shown above leading to their representation in the tree chart [Fig 5.2.8], where the 1st truth (suffering) comes to be understood in terms of the 5 aggregates or the 6 sense-bases, and the 4th truth in terms of the noble eightfold path.

5.2.8.3 Over time, as more teaching sets arose in the Buddha’s teaching, 7 of them became prominent. Although they often feature separately as the respective themes of their own suttas, they are often mentioned together as a set of 7 teachings—”the 7 sets”—without any special name given to them, except by way of the totality of their respective items as the 37 states conducive to awakening (satta,tiṁsa bodhi,pakkhiya,dhamma), or simply, “the 37 limbs of awakening.”290 4 of these 7 sets can be sporadically found as part of a tree diagram [Fig 5.2.8]. The 7 sets, constituting the 37 limbs of awakening, are prominent enough to be mentioned in the parable of the great ocean—highlighting the key qualities of the Dharma—as being comparable to the 7 treasures found in the ocean depths.291 The passages run thus:

Pahārāda, the great ocean has many treasures, diverse treasures. In it there are these treasures, that is to say: pearl, crystal, beryl, conch, quartz, coral, silver, gold, ruby, cat’s-eye292— so, too, Pahārāda, this Dharma-Vinaya has many treasures, various treasures. Here, there are these treasures, that is to say [the 37 limbs of awakening in 7 sets]: 1. the 4 focuses of mindfulness, catu satipaṭṭhāna 2. the 4 right strivings, catu samma-p,padhāna

289 See Sambādh’okāsa S (A 6.26), SD 15.6. 290 Bodhi,pakkhiya,dhamma: see SD 10.1; Pārileyya S (S 22.81,11), SD 6.1; SD 9 (10.3); Sakul’udāyī S (M 77,15-21), SD 6.18. On how these teachings work together in practice, see SD 45.18 (2.7, 2.8.4.2). 291 This pericope, with 8 parables + 8 qualities, occurs 4 times: Pahārāda S (A 8.19/4:197-204) + SD 45.18 (1.2), Buddha speaks with asura leader, Pahārāda; (Aṭṭhaka) Uposatha S 1 (A 8.20/4:204-208) Buddha uses these parables to stress that Pātimokkha conclave should be properly observed; he lists only the pericope; (Aṭṭhaka) Uposatha S 2 (U 51-56) is identical with the last except it has an udāna (Tha 447) [Tha:N 196 n447]: this is prob the latest version; Pātimokkha Ṭhapana Khandhaka (Cv 9.1.2-4 @ V 2:236-240), too, follows this 2nd version except that certain words are moved to the end. Of these, A 8.19 is the shortest and clearly the oldest: Frauwallner agrees (1956:147 f). 292 On these 10 “treasures,” see SD 45.18 (2.7.2). http://dharmafarer.org 61 SD 58.1 The Oral Tradition of the Early Buddhists

3. the 4 paths to spiritual success, catu iddhi,pādā 4. the 5 spiritual faculties, pañc’indriya 5. the 5 spiritual powers, pañca bala 6. the 7 awakening-factors, satta bojjhaṅga 7. the noble eightfold path.293 ariya aṭṭh’aṅgika magga (A 8.19,17(7)/4:203), SD 45.18

5.2.8.4 Other lists would evolve out of the practice of taking an item or aspect of a list and explaining it by way of a carefully structured “analysis” (vibhaṅga), which, in turn, can then be conveniently summed up numerically as a list. Thus, an “analysis” of the eightfold path does not explicitly state that right thought consists of “the 3 thoughts,” rather it simply gives what is, in effect, a threefold analysis of right thought. In fact, an explicit list of “3 wholesome thoughts” is found in the Saṅgīti Sutta (D 33).294 Other numerically structured analyses, such as the one based on the 3rd focus of mindfulness—the contemplation of mind (cittânupassana)—does not appear evolved into a list of its own. We see here a difference between what a list teaches, and what each teaching does or becomes in practice. The number of breaths we take in meditation will not be useful in any such list.

5.2.8.5 By way of summary, we can make two observations. First, just by remembering the list of the 4 noble truths, we have access into a vast variety of sutta teachings. In other words, we can see how the lists operate as a basic mnemonic device enabling us to remember a lot of teachings. There appear to be 3 main ways in which the lists do this:

(1) a list subsumes another list, for example, the list of the 4 noble truths subsumes the list of the 5 aggregates under the 1st truth; (2) one list may be substituted for another in a given context, for example, under the 1st aggregate, the 4 elements many be substituted for the 5 sense-faculties; (3) one list may suggest another list by association in important sutta contexts, for example, the 4 dhyanas suggest the overcoming of the 5 hindrances.

5.2.8.6 Although these lists are mnemonic tools—they help us remember the teachings—there is a higher purpose for remembering, even learning, such teaching. As we become more familiar with these lists, we begin to see how they fit together to form a pattern. They interconnect to form structures of words, wisdom and ways that are “beautiful in the beginning, beautiful in the middle, beautiful in the end.” Suppose we are giving teachings on the 4 noble truths: we have a number of options. We may give a comprehensive exposition, expanding in full all the subdivisions, and secondary and tertiary lists shown in Fig 3.2.8. Or, we may present the 4 truths simply yet precisely. Yet, again, we may decide to focus on only one branch, ending up with an expanded talk on, say, the 1st focus of mindfulness (contemplation of body)295 or the abandoning of the 5 hindrances and the attainment of the 1st dhyana.296 In such cases the underlying basis of the 4 truths need not be explicitly mentioned, serving only rudder or anchor for our cruise of the Dharma.

293 Evam evaṁ kho pahārāda ayaṁ dhamma,vinayo bahu,ratano aneka,ratano, tatr’imāni ratanāni, seyyath’idaṁ cattāro sati’paṭṭhānā cattāro samma-p,padhānā cattāro iddhi,pādā pañc’indriyāni pañca balāni satta bojjh’aṅgā ariyo aṭṭh’aṅgiko maggo. 294 D 33,1.10(6)/3:215. 295 See Kāya,gata,sati S (M 119/3:88-99), SD 12.21. 296 See Sāmañña,phala S (M 2,68-76/1:71-74), SD 8.10.

62 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

5.2.8.7 We must imagine how the lists not only help mechanical memorization, rote learning, but are also a kind of flowchart for the composition, or better, re-composition, of a sutta or teaching. However, there is latitude for variation within the spirit of the Dharma. They present various paths and themes that the reciter or redactor can choose to follow and expand as he sees appropriate or as appropriate to the circumstance. The matrix of interconnecting lists provides a form or structure within which he may improvise. He must, of course, know both contents and structure well, as clear from our discussion of āgama [5.2.1]. The early reciters were adepts; the adepts were the reciters. They knew the texts and structure well, and were well endowed with memory and teaching skills. They never lost their way since they have reached the destination.

5.2.8.8 The Pali canon, as we know it today, is a fixed literary text. Clearly, it was not always so. Given the prevalence of interlinking lists, we can easily see how there may be one version of a sutta mentioning the 4 focuses of mindfulness as a bare list, another mentioning them with a brief exposition, and yet another that goes on to give a full exposition. This trend is, in fact, what we see in the Pali canon itself. When we understand this natural latitude that pervades the early oral tradition (5.2.8.7], we see no need for any doubt about an “original” version of a sutta, especially in the context of the Pali canon. To a great and helpful extent, we have the benefit of this latitude in a comparative research between the Pali Nikāyas and the Chinese Āgamas.

5.2.8.9 At this point, we must turn our attention to a characteristic feature that we see in the Pali suttas but much less so in the Chinese Āgamas. This is the peyyāla or “abridgement for repetition” (or simply, repetition) the trend of which we have already mentioned [5.2.8.7]. The peyyāla sections of the Saṁyutta and the Aṅguttara Nikāyas are especially interesting in this respect. For, even as we have them, they indicate a basic pattern or formula that is to be applied to various teachings in succession. The result is a text with quite radical abridgements. Indeed, from the manuscripts and editions we have today, it is not always clear just how or how much we must expand the material to get the “full” text, assuming that this was its original intention. Clearly, a certain freedom is intended here. Hence, these Saṁyutta and the Aṅguttara peyyāla sections must be read more like guidelines for oral recitation and composition than a fixed literary text. Yet, this is not a “random” or ad hoc improvisation of text composition. We have clear guidelines from preceding texts in preceding sections or other Nikāyas. For the neophytes, these “set texts”—those in the preceding sections and other Nikāyas—are a vital guide to vision and direction of such peyyālas. This latitude attests to the reality that they reflect a “living text,” that the Pali texts (despite being the “early Bud- dhist texts”) are not fixed scripture, but is veritably an open spiritual tradition in the hands and eyes of the adept, and those who aspire for the path of these adepts, the arhats. In closing this section, let us remind ourself [5.2.8.1] that the oral tradition suggests the prevalence of the “reciter’s memory” amongst the early monastics who knew the Dharma both “in the letter and in the spirit.” If we are Dharma-spirited practitioners who are unable to immediately recall a sutta or its teachings, we will always be guided by the Dharma’s spirit, like the adepts before us. The vision is the same, only for us, we still need to move up the path.

5.2.9 Numerical lists

5.2.9.1 Early Buddhist literature contains, then, a great number of lists. Clearly, certain lists are more significant than others; some lists occur perhaps in only one context, whereas others crop up again and again. In such circumstances it is hardly surprising to find, at a relatively early date, the Buddhist tradition itself focusing on particular groups of lists and drawing up composite lists, that is, lists of lists. Probably http://dharmafarer.org 63 SD 58.1 The Oral Tradition of the Early Buddhists one of the earliest of such composite lists is the 7 sets of teaching [5.2.92], that later came to be collectively known as “the 37 limbs of awakening” (bodhi,pakkhiya,dhamma). Well known among these lists are also the 5 aggregates (pañca-k,khandha), the 12 sense-bases (dvā- dasa dhātu), and the 18 elements (aṭṭhārasa dhātu), collectively called khandh’āyatana,dhātu.297 Other early lists of lists include the 2 famous set of questions, the Kumāra,pañha and the Mahā,panha [5.2.9.3], and, of course, the more extended discourses, the Saṅgīti Sutta (D 33) and the Das’uttara Sutta (D 34).

5.2.9.2 The simplest use of a numbered listing—or numerical association and numerical progression, to be exact—is found in a basic list of 10 questions, said to be first recorded as being used as the going-forth procedure for the precocious 7-year-old novice, Sopāka, who is already an arhat.298 It is also said to have been taught to the 7-year-old novice Rāhula.299 The Nigaṇṭha Nāta,putta Sutta (S 41.8) relates how the householder Citta asks the Jain teacher, Nātaputta, these questions, but he is unable to answer them.300

(1) Appropriately called “the boy’s questions,” Kumāra,pañha (Khp 4/2), this brief questionnaire runs as follows:

Ekaṃ nāma kiṃ? Sabbe sattā āhāra-ṭ,ṭhitikā. What is one? All beings are sustained by food. Dve nāma kiṃ? Nāmañ ca rūpañ ca. What is two? Name and form. Tīṇi nāma kiṃ? Tisso vedanā. What is three? The 3 feelings. Cattāri nāma kiṃ? Cattāri ariya,saccāni. What is four? The 4 noble truths. Pañca nāma kiṃ? Pañc’upādāna-k,khandhā. What is five? The 5 aggregates of clinging. Cha nāma kiṃ? Cha ajjhattikāni āyatanāni. What is six? The 6 internal sense-bases. Satta nāma kiṃ? Satta bojjhaṅgā. What is seven? The 7 awakening-factors. Aṭṭha nāma kiṃ? Ariyo aṭṭhaṅgiko maggo. What is eight? The noble eightfold path. Nava nāma kiṃ? Nava satt’āvāsā. What is nine? The 9 abode of beings. Dasa nāma kiṃ? Dasa-h-aṅgehi samannāgato What is ten? One endowed with the 10 limbs “arahā’ti vuccatîti. is called an arhat.301 (Khp 4/2), SD 3.10 (1.2.1; 2)

(2) Besides the Kumāra,pañha (Khp 4), there are the Mahā,pañha Sutta 1 (A 10.27) and its variant, the Mahā,pañha Sutta 2 (A 10.28).302 Each of them gives 10 teachings, simply listed from 1 to 10. This set of 10 teachings clearly are those to be recited, studied in detail and remembered by the neophyte. On account of their variance, it is likely they preceded the 7 sets; or, they arose around the same time, when the teachings began to take more formal shape as they are disseminated to a growing monastic system of unawakened renunciants and the lay community.

297 See Tha 1255; Thī 43, 69, 103; Ap 563; Nm 45. 298 Tha 448-486; KhpA 75 f; SD 3.10 (1.2.1); SD 45.16 (2.4). 299 MA 3:126; AA 1:258. 300 S 41.8/4:299 f (SD 40a.7). 301 These are the tenfold rightness (sammatta), viz, the eightfold path—right knowledge (sammā ñāṇa) and right freedom (sammā vimutti): SD 10.16 (1.8-10). 302 Mahā,pañha S 1 (A 10.27/5:48-54), SD 85.15; Mahā,pañhā S 2 (A 10.28/5:54-59), SD 85.16. For a Chin parallel to the former, see EĀ 46.8 (&T2.778b17), which, however, diverges in the foll: the 5 faculties, the 6 principles for communal harmony, the 10 kinds of mindfulness (the 6 recollections, mindfulness of the body, of the breath and of peace). It however gives explanations for the items in each number. A 10.27 mentions 3 categories for each number—question (pañha), teaching (uddesa) and explanation (veyyākaraṇa) but gives only the question and teaching.

64 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

(3) The 7 sets [5.2.8.3] feature sporadically in the Kumāra,pañha, which gives the 7 awakening-factors and noble eightfold path for the numbers 7 and 8, but the 4 noble truths and 5 aggregates of clinging for 4 and 5; the variation in the Mahā,pañha Sutta 2 gives the 4 focuses of mindfulness and the 5 faculties. [Comparative table below.] Here is a comparative table summarizing the teachings of each of the lists in the 3 texts:

Kumāra,pañha (Khp 4) Mahā,pañha 1 (A 10.27) Mahā,pañha 2 (A 10.28) 1. all beings subsist on food the same the same 2. name-and-form the same the same 3. the 3 kinds of feelings the same the same 4. the 4 noble truths the 4 kinds of food the 4 focuses of mindfulness 5. the 5 aggregates of clinging the same the 5 faculties 6. the 6 internal sense-bases the same the 6 elements of escape303 7. the 7 awakening-factors the 7 stations of consciousness the same as Khp 4 8. the noble eightfold path the 8 worldly conditions the same as Khp 4 9. the 9 abodes of beings the same the same 10. the arhat’s 10 limbs the 10 unwholesome courses of the 10 wholesome courses of karma karma

Clearly, the Kumāra,pañha—which includes the 4 noble truths, the 7 awakening-factors, the noble eightfold path and the arhat’s 10 limbs—are for the training of a renunciant as its origin story attests. The Mahā,pañha 1 relates how its 10 questions are taught by the Buddha in response to a challenge by some outside wanderers who claim that they know “the all” (sabbaṁ dhammaṁ). From the last 3 items of the Mahā,pañha 1 and 2, it is likely that they are learning decads for the laity: the former for general practice, the latter for those who meditate. Significantly, too, the Mahā,paṇha 2 is taught by a wise nun to the laity [5.2.9.3].

5.2.9.3 The Mahā,pañha Sutta 2 relates how lay followers from Kajaṅgala304 approach a nun there — only known as “the nun from Kajaṅgala” (kajaṅgalikā bhikkhuṇī)—and ask her to answer in some detail about the 10 great questions (mahā,pañha) which the Buddha has spoken on only briefly. She gives a most significant reply: “Avuso, I have neither heard nor learned this (the great questions) in the Buddha’s presence, nor have I heard or learned this in the presence of esteemed monks. However, listen and give attention as I explain what it means to me.”305 Her answers are recorded in the Mahā,paṇha Sutta 2 (A 10.28), which when brought before the Bud- dha, receives his full approval. He praises the nun from Kajaṅgalā as being very wise, and that he would himself have given the same answers.306 This is a good example of how the āgama tradition works. Even

303 Nissaraṇīyā dhātuyo, viz: (1-3) dhyana through the 4 divine abodes, (5) the signless freedom of mind, (6) up- rooting of “I am” conceit”: (Chakka) Mettā S (A 6.13/3:290-292), SD 60.2. For a different set of 6: (Pañcaka) Nissa- raṇīya S (A 5.200/3:245 f), SD 55.17. 304 In the Buddha’s time, Kajaṅgalā or Gajaṅgalā (Be) was a prosperous market-town, “where provisions are easily obtained (dabba,sambhārā sulābhā, J 4:310). forming the eastern boundary of Middle Country (majjhima,desa), the holy land. Beyond it was Mahā,sālā (V 1:197; DA 1:173; MA 2:200; AA 1:97, 2:36; kHPa 132; J 1:49; SnA 1:300; ItA 2:81; ApA 53). See SD 52.1 (2.2.1.10+18). See DPPN: Kajaṅgala. 305 Na kho pan’etaṁ avuso bhagavato sammukhā sutaṁ sammukhā patiggahītaṁ; ba ca mano,bhāvanīyānaṁ bhikkhūnaṁ sammukhā sutam sammukhā patiggahītaṁ; api ca yathā m’ettha khāyati, taṁ suṇātha sādhukaṁ manasikarotha bhāsissāmîti (A 10.28,3/5:55), SD 85.16. 306 A 10.28/5:54-59 (SD 85.16); KhpA 8, 83, 85. For a Skt account: Avś 2.78 (where the town is called Kacaṅgala). http://dharmafarer.org 65 SD 58.1 The Oral Tradition of the Early Buddhists though the nun, living in a remote part of ancient India, remote from the well known centres of Buddhism, was able to answer the “great questions” to the Buddha’s approval.

5.2.9.4 Rather similarly, various of the 7 sets feature in the numerical system of the Das’uttara Sutta (D 34): the 4 focuses of mindfulness, the 7 awakening-factors and the noble eightfold path as 4, 7 and 8 dharmas respectively to be cultivated (bhāvetabba); the 5 faculties as the 5 states concerned with distinction (visesa,bhāgiya),307 though a Chinese translation of the Daśottara Sūtra, treats them as “to be cultivated” [5.2.9.6]. The broader numerical method of the Saṅgīti Sutta (D 33) features all the 7 sets. In the compilation of these composite lists, 2 methods seem to have been used:

(1) by a convenient mnemonic summary of aspects of the Dharma, that is, a grouping of closely related (apposite) lists (for example, the 7 sets of teachings); (2) a more general summary following the principles of numerical association (bringing together different lists that all comprise the same number of items) and/or numerical progression (taking a list comprising 1 item; then, a list comprising 2 items, and so on up to 10 or 11 items).

Method (2) is the one used in the Saṅgīti Sutta (D 33), the Das’uttara Sutta (D 34), the Kumāra,pañha Sutta (Khp 4), the Mahā,pañha Sutta 1 (A 10.27) ) and Mahā,pañha Surra 2 (A 10.28). In fact, both these methods are found in the classic collections of the Saṁyutta Nikāya and the Aṅguttara Nikāya. These composite lists are no doubt intended to function as succinct compendia of the Dharma, but at the same time they also appear to be regarded as representing a kind of essence of the Dharma, distilling them into lists in the suttas, so that these lists are seen “as laying the Dharma bare and revealing its inner workings.”308

5.2.9.5 Thus, the various composite lists may be viewed as different ways of seeing the structure at the very heart of the Dharma. In undertaking the task of compiling these composite lists, the early Bud- dhist sutta redactors seems to have felt that it was not quite enough simply to list the lists one after another; for, as we have already seen [5.2.8.1], to understand the lists is to know where they fit together in the fullness of the Buddha Dhamma. The Saṅgīti Sutta (D 33) puts together lists with the same number of teachings, starting with “ones” and ending with “tens,” which seems to simply be nothing more than a simple mnemonic device for remembering various teaching sets. Its simple numerical list of lists seems to be a preliminary exercise, which is more developed in the Das’uttara Sutta (D 34) which immediately follows it. Both suttas arrange their teachings using the principles of numerical association and progression [5.2.9.4]. The latter, however, builds up a whole series of lists, up to 100. In either Sutta, their structure shows just how each teaching fits into the Dharma as a whole.

5.2.9.6 The Daśottara Sūtra, which survives in Buddhist Sanskrit and Chinese translations, is a Sanskrit parallel to D 33 and 34 [5.2.9.5]. In addition to the lists of the 2 Pali suttas, it also gives various alternative lists.309 For this reason, argues Gethin, “this seems to me a very good illustration of why we should not think in terms of an ‘original’ or ‘correct’ version of such a text.”310 What we have here, then, is a mnemonic technique and system of arrangement built around numerical association and progression [5.2.9.4]. These arrangements go beyond mere rote learning, revealing an

307 See Dhamma,cetiya S (M 89,12 + 18), SD 64.10; Bhikkhuṇī Vāsaka S (S 47.3 + 10), SD 24.2; Ānâpāna,sati S (M 118,2+6), SD 7.13. 308 Gethin, 1992b:157. 309 See J W de Jong, “The Daśottarasūtra,” 1966:3-25; repri 1979:251-273. 310 Gethin, 1992b:157 f.

66 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists inner structure of awakening truths. This can only happen, of course, when we know what we are doing, even as we improvise as appropriate so that the teaching is well understood and benefits the audience and posterity. This is called “skilful means” (upāya).311

5.3 MĀTIKĀ AND MĀTIKĀ,DHARA (Gethin)312

5.3.1 The mātikā in the Abhidhamma

5.3.1.1 Towards the close of the Nikāya period, we see an interesting term being used in the early Buddhist texts: the term mātikā. In the 4 primary Nikāyas and the Vinaya Piṭaka, this term is character- istically found as mātikā,dhara, the last component of a well known compound.313 This compound always occurs as the 3rd component in the sequence dhamma,dhara vinaya,dhara mātikā,dhara that forms part of a stock description of an accomplished disciple, one who is “learned, deeply learned, mastered attainment, Dhamma-expert, Vinaya-expert, Matrix-expert (mātikā,dhara).” [5.2.1.2]. We also find the term mātṛkā similarly employed in Buddhist Sanskrit sources.314 But what exactly is a mātikā? Buddhaghosa (5th century CE) understands mātikā in the context of mātikā,dhara as referring to the 2 Pātimokkhas (one for the monks, the other for the nuns), or the bare code of rules for fully ordained monks and nuns extracted from their Vinaya context in the Sutta,vibhaṅga, dealing with the monastic rules and their case histories.315 The word mātikā is certainly used in this sense in the Commentaries, probably from a relatively early date.316 However, such an interpretation appears too specific, even anachronistic, and is not supported by the evidence found elsewhere in the texts.

5.3.1.2 The well known sequence, dhamma vinaya mātikā, clearly corresponds to the sutta vinaya abhidhamma in the accounts of the 1st Buddhist council in the Pali Vinaya,317 corroborated by surviving Chinese and Tibetan parallels, which say that after Ānanda had recited the Sūtras, and Upāli the Vinaya, Mahākāśyapa recited the mātṛkās.318

311 See Upāya, skilful means, SD 30.8. 312 This section is a summary (with occasional comments) of R Gethin, “The Mātikās,” 1992b:158-167, which accounts the probable rise of the Abhidhamma. 313 Mātikā, however, is used in 2 contexts in Vinaya, in the ordinary figurative sense of “source” or “origin” (PED, sv mātikā): there are “8 grounds for the withholding of kaṭhina (privileges)”(aṭṭha mātikā kaṭhinassa ubbhārāya) and “8 sources for the making of a robe“ (aṭṭha mātikā cīvarassa uppādāya); see V 1:255, 309, 3:196, 199, 5:136, 172-174. Cf Mūlasarvāstivāda Vinayavastu, Gilgit Manuscripts, ed N Dutt. vol 3, part 2, Srinagar, 1942: 161: aṣṭau mātṛkā-padāni kaṭhinôddhārāya samvartante (Edgerton mistakenly tr as “8 summary points,” see BHSD sv mātṛkā). The word is also used (apparently only once in Nikāyas, once in Abhidhamma) to mean “water-course” or “channel” (A 4:237,11; DhsA 269,26). 314 See BHSD: mātṛkā. 315 MA 2:189, 3:382. 316 See K R Norman, Pali Literature 1983:96, 126; the Parivāra seems to use the term in this sense, too (V 5:86). 317 Cv 11 (V 2:284-293). Although this account only mentions dhamma,vinaya, without mentioning abhidhamma, cf the closing verses, where it is said: “the Conqueror’s disciples recited the 3 Piṭakas” (piṭakaṁ tīṇi saṅgītiṁ akaṁsu jina,sāvakā, V 2:293,5). This seems to mean that the Abhidhamma was added by Asoka’s time when the Pali canon was closed. 318 See esp Mūlasarvāstivāda Vinaya accounts (Tib and Chin) and some versions of Aśokāavadāna: W W Rockhill, The Life of the Buddha 1907:60: J Przyulski, Le concile de Rājagṛha 1926:45; cf J Bronkhorst, “Dharma and Abhidhar- ma,“ BSOAS 48, 1985:305-320 (esp 320). http://dharmafarer.org 67 SD 58.1 The Oral Tradition of the Early Buddhists

It thus seems possible that mātikā must be the early name for the Abhidhamma.319 Here, Gethin warns us: “Although in what follows I certainly do not wish to deny that a relationship exists between the mātikās and the development of the Abhidhamma, it seems to me that to suggest any simple equivalence of the two terms must be regarded as a misleading simplification.” (1992b:158) The mātṛkās that Mahā Kāśyapa is said to have recited comprise the 7 sets beginning with the 4 focuses of mindfulness, along with a number of other lists. This is one of the reasons that led A K Warder to see in this list the basis of the “original” or primary mātikā of the Abhidhamma.320 However, before undertaking a search for the original Abhidhamma mātikā, reminds Gethin, it is worth considering further the actual use of the term in the Pali sources. (1994b:158)

5.3.1.3 Apart from its use in the compound mātikā,dhara (where we simply do not know precisely what mātikā refers to), the most extensive use of the term mātikā in the Pali texts is in the Abhidhamma Piṭaka. Its use here is quite specific and probably constitutes the earliest evidence for the technical application of the term. In the first place mātikā is used to describe the list of 32 “triplets” (tika) and 100 “couplets” (duka) set out at the beginning of the Dhamma,saṅgāṇī.321 Each triplet comprises 3 categories for classifying Dhammas (mental states); each couplet comprises 2 such categories. Essentially, the Dhamma,saṅgaṇī is an exercise in expounding this mātikā (more later). The mātikā of the triplets and couplets is also employed by 3 other canonical Abhidhamma works, namely, the Vibhaṅga, the Dhatu,kathā, and the Paṭ- ṭhāna. The Abhidhamma section of the Vibhaṅga on “the analysis of the modes of conditioning” opens with a mātikā that indicates 144 variations of the dependent-arising formula that are built up systematically around 16 basic variations (arranged in groups of 4), which are each subject to a further 9 variations; 322 the exposition that follows begins to apply each variation in turn to the different kinds of consciousness (citta) distinguished in the Dhamma,saṅgaṇī.

5.3.1.4 Like the Saṁyutta and the Aṅguttara Nikāyas, the Abhidhamma texts are full of abbreviated repetitions (peyyāla). In “the analysis of meditation” (jhāna), the Suttanta section opens with a mātikā.323 This is a rather untypical mātikā; it is made up of stock Nikāya formulas describing the attainment of the 4 form dhyanas and 4 formless attainments. The exposition that follows consists of a straightforward word-commentary. Two further mātikās occur at the beginning of “the analysis of the items of knowledge” and “the analysis of minor items,” respectively.324 Both these mātikās consist of a schedule compiled (like the Saṅgīti Sutta, the Das’uttara Sutta, the Aṅguttara Nikaya, etc) according to a principle of numerical progression from 1 to 10. All relevant “ones” are listed, then all relevant “twos,” and so on until we reach “tens,” the exposition that follows then provides a detailed explanation of all items.

319 See PED, sv mātikā; BHSD, sv mātṛkā; Lamotte, History of Indian Buddhism 1988a:149 [Histoire 1958:64]; Nor- man, Pali Literature 1983b:96. 320 [39] Warder, “The Matika,” in Moh 1961:ix-xxvii (esp xx). 321 [40] Dhs 1-7. This is the abhidhamma,mātikā; there is appended a suttanta,mātikā consisting of a further 42 couplets (Dhs 7 f). 322 [43] Vbh 138-143. 323 [44] Vbh 244 f. 324 [45] Vbh 306-318, 345-349.

68 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

5.3.1.5 The Dhātu,kathā opens with a rather more complex mātikā that falls into 4 parts:

(1) 14 pairs of categories of analysis; (2) 22 sets of items to be analyzed; (3) an indication of the path the analysis is to follow; (4) the 22 triplets and 100 couplets of the mātikā from the Dhamma,saṅgaṇī, which are also to be analyzed.

The rest of the Dhātu,kathā takes the form of a relatively concise and restrained working out of this mātikā. The Puggala,pannatti opens once more with a straightforward mātikā that arranges the headings to be discussed in the text according to the system of numerical progression from 1 to 10. The Kathā,vatthu and the Yamaka do not have explicit mātikās, although once again later tradition sees fit to describe both the underlying list of discussion points in the Kathā,vatthu and the aggregate of the ten lists that form the basis of the Yamakas 10 chapters as mātikās. 325 In all, there are 8 explicit mātikās in the texts of the Abhidhamma Pitaka: 2 in the Dhamma,saṅgaṇī, 4 in the Vibhaṅga, one each at the beginning of the Dhātu,kathā and the Puggal,apannatti. 326

5.3.2 The mātikā outside the Abhidhamma

5.3.2.1 The term mātikā is similarly employed outside the Abhidhamma Pitaka in the Patisambhidā,- magga, the 12th book of the Khuddaka Nikāya, which consists of 30 “talks” (kathā) on various topics. The themes selected and the arrangement of the text are distinctive. The opening “talk on knowledge” starts with a mātikā.327 This lists 73 kinds of knowledge that are then explained in the “talk” that follows. As A K Warder notes,328 of the 30 talks,the first is by far the longest (constituting about 1/3 of the whole text), and within this talk only the first of the 73 kinds of knowledge gets the full treatment. The Patisambhidā,magga opens with a mātikā and closes with a “talk on a mātikā.” 329 The mātikā in question consists of a series of somewhat miscellaneous terms that appear, from the subsequent exposition, to be intended to constitute 15 divisions. Again, our text is radically abbreviated; great formulas and long lists employed earlier in the work are to be inserted to work out the exposition in full.

5.3.2.2 It would appear, then, that a mātikā can be any schedule or table of items or lists—but especially one built up according to a system of numerical progression—that acts as a basis for further exposition. The commentarial application of the term to the bare list of Vinaya rules hardly stretches this understanding.330 At this point, it is worth considering how the Sanskrit equivalent, mātrkā, is used beyond the confines of Buddhist literature. A secondary formation derived from the ordinary word for “mother” (mātṛ), mātṛkā (cognate with English “matrix”) is apparently used in the first place, again simply to mean “mother” and, in addition, “grandmother.” Hence, it also figuratively means “source” or “origin” in general.331

325 See DhsA 4; KvuA 7; Moh 3, 257, 278; cf Norman, Pali Literature, 1983:96, 105. 326 Gethin: Norman’s reference (Pali Literature, 1983:106) to a mātikā at the beginning of the Paṭṭhāna appears mistaken (1994b n47). 327 Pm 1:1-3. 328 Pm:Ñ 1982:xviii f. 329 Pm 2:243-246. 330 Cf Norman, Pali Literature, 1983:96. 331 SED sv mātṛkā. http://dharmafarer.org 69 SD 58.1 The Oral Tradition of the Early Buddhists

The Sanskrit senses of the term does not seem to apply in any of the Pali and Buddhist Sanskrit use of mātikā/mātṛkā [5.3.1.1]. Translators of Buddhist texts have often taken the word to mean something like “summary” or “condensed content.” Yet, a mātikā is not so much a condensed summary as the seed from which something grows. “A mātikā is something creative—something out of which something further evolves. It is, as it were, pregnant with the Dhamma and able to generate it in all its fullness.”332

5.3.2.3 Kassapa of Coḷa (fl c1200 CE) explains mātikā in his Moha,vicchedanī, a commentary on the mātikās of the Abhidhamma, as follows:

In what sense is it a mātikā? In the sense of being like a mother. For a mātikā is like a mother as a face is like a lotus. For, as a mother gives birth to various different sons, and then looks after them and brings them up, so a mātikā gives birth to various different dharmas and meanings, and then looks after them and brings them up so that they do not perish. Therefore, the word mātikā is used. For in dependence on the mātikā, and by way of the 7 treatises beginning with the Dham- ma,saṅgaṇī, dharmas and meanings without end or limit are found as they are spread out, begot- ten, looked after and brought up, as it were, by the mātikā.333 (Moh 2)

Kassapa goes on to explain that if the 7 canonical Abhidhamma treatises were expanded in full, each one would involve a recitation without end or limit (anantâparimāṇ.bhāṇa,vāra). He then concludes:

Thus, the word mātikā is used because of the begetting, looking after and bringing up of dharmas and meanings without end or limit like a mother. And looking after and bringing up here are to be understood as the bringing together and preserving of the neglected and hidden meanings of the texts, having distinguished them by following the mātikā.334 (Moh 3)

5.3.2.4 We can perhaps sum up by saying that mātikās contain the building blocks for constructing an exposition or text. But they are sort of “magical building blocks”; when combined and used in various ways they can create a palace that is much larger in extent than the sum of the parts. If the lists and schedules that we have been considering are mātikās, then someone who is mātikā,dhara, “one learned in the mātikās” or the matrix-expert, is presumably someone who knows these building blocks and similar lists. More importantly, he also knows what to do with them. in other words, he knows how to expand them and expound the Dharma from them. One who is mātikā,dhara, then, is not simply someone who can spout endless lists of lists learnt by rote, but one who improvises and creates through the medium of these lists. Such a matrix-expert is, in fact, a disciple (at least a streamwinner, usually an arhat), who has teaching abilities, and can spontaneously come up with teachings and teaching methods, on account of his path-attainment.

332 Gethin 1992b:161. 333 Moh 2: ken’atthena mātikā. Mātu,samatthena. mātā viyâti hi mātikā yathā padumikam mudhan’ti. yathā hi mātā nānā,vidhe putte pasavati te pāleti poseti ca evam ayaṁ pi nānā,vidhe dhamme atthe ca pasavati te ca avi- nassamāne pāleti poseti ca. tasmd mātikā’ti vuccati. mātikaṁ hi nissaya dhamma,saṅaṇi,ādi,satta-p,pakaraṇa- vasena vitthāriyamānā Ananta,parimānā dhammā atthā ca tāya pasūta viya pālitā viya positā viya ca labbhanti. 334 Moh 3: evam anantâparimāṇānaṁ dhammānaṁm atthānañ ca pasavanato pālanato posanato ca mātā viyâti mātikā’ti vuccati. pālana-posanañ c’ettha pamuṭṭhānaṁ viraddhānañ ca pāḷi.atthānaṁ mātikā’nusārena sallak- khetvā samānayanato rakkhanato ca veditabbaṁ.

70 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

5.3.2.5 Gethin again warns us:

“All this certainly suggests some relationship between the mātikā and development of the Abhidham- ma, but we must, I think, be wary of understanding the earliest mātikās in terms of a distinct and separate body of literature existing alongside the Vinaya and Sutta Pitakas. Rather, the Abhidhamma would appear to evolve out of an already developed practice of taking a list or combination of lists, and then expanding it to produce an exposition.”335

5.3.3 The rise of Abhidhamma

5.3.3.1 When we look at the scholastic nature of the Abhidhamma lists and theoretical taxonomy and exegesis, they betray a contrived tendency to measure or fix the fluid and living Dharma expressions of the suttas into technical terms that seem to stand on their own. In other words, we see a nascent hint of the rise of Buddhist dogmatics that we call Abhidhamma. Clearly, this cannot be the work of awakened sutta redactors—certainly not the Buddha—but the genius of a learned group of brilliant minds who had the benefit of closeness (propinquity) with the great saints, the noble individuals. The Abhidhamma is certainly the work of brilliant minds, but unawakened hearts: they are the scholastics, those who let the word overtake the spirit of the Dharma.336 We have an instructive sutta account of such monks who seem to be driven by the word rather than the heart of the Dharma. The (Chakka) Mahā Cunda S (A 6.46) is related by Mahā Cunda to the monks at Sahajāti337 regarding antagonistic conduct between monks who are Dharma specialists (dhamma,yogā) and those who are meditators (jhāyī).338 We probably would be right to surmise that this is an episode during a time when the monasteries are large and well established in India.339 It is likely that such Dharma specialists are those who initiated or contributed to the rise of the Abhidhamma tradition.

5.3.3.2 Toward the end of the Nikāya period, the way in which lists were being used approaches closer to the more formalized Abhidhamma use of mātikās. Appropriately enough, the contents of both the Saṅgīti Sutta (D 33) and the Das’uttara Sutta (D 34) are shown not as coming from the Buddha himself but from Sāriputta, who is strongly associated with the Abhidhamma.340 In fact, the Sarvastivadins include the

335 Gethin adds: “This is a practice that, in principle, goes right back to the beginnings of Buddhist literature, grad- ually becoming more formalized as the body of material increased in size and certain lists acquired a special significance.” 336 We are reminded of the “professional Buddhists” of our time, who, by their professional abilities, try to re- define Buddhism or present it in modern “psychology,” “management” and so on. 337 Sahajāti is a market-town of the Ceti, east of the river Yamunā, next to Kuru country, in the far west (V 4:108 f; J 1:360 f). It was one of the 16 great countries (mahā,janapada) of Buddha’s India [SD 4.18 App; SD 9 (16): map (16.3)]. See DPPN: Ceti (1). 338 Jhāyī here clearly simply means “meditator,” and is unlikely to be “dhyana-meditators,” for at least 2 reasons: they seem to be living in a large community, perhaps together with the Dharma-speakers (dhamma,kāthikā, AA 3:379). Moreover, since they are engaged in such conflict, it is unlikely they have attained dhyana anyway. Moreover, Mahā Cunda’a lesson of the account is highly significant. 339 (Chakka) Mahā Cunda S (A 6.46/3:355 f), SD 4.6. 340 Attha,sālinī (DhsA) tells the famous story of Kassapa Buddha’s time, of the cave bats (incl Sāriputta in his past life) listening to 2 monks reciting the Abhidhamma (DhSA 1:16 f). It adds: “The textual order of the Abhidhamma originated with Sāriputta; the numerical series in the Great Books [pakaraṇa] was also determined by him.” (abhi- dhammo vācanā,maggo nāma sāriputta-t,thera-p,pabhavo mahā,pakaraṇa,gaṇana,cāro pi theren’eva ṭhapito, DhsA 1:17,12-14; DhsA:P, 21, Pe Maung Tin’s tr). http://dharmafarer.org 71 SD 58.1 The Oral Tradition of the Early Buddhists

Saṅgītiparyāya, a text based on their recension of the Saṅgīti Sutta, among their canonical Abhidharma works. The works of the canonical Abhidhamma, then, in part, are to be seen as the result of a process of drawing up mātikās and evolving from a similar process nascent in the suttas. We see the first signs of such works by masters of the mātikās, the matrix experts, in such early texts as the Saṅgīti Sutta and the Das’uttara Sutta, the Kumāra,pañha Sutta, the 2 Mahā,pañha Suttas [5.2.9], and also in the suttas of the the Saṁyutta and the Aṅguttara. [5.2.9.4]

5.3.3.3 The Aṅguttara employs the same system of numerical arrangement, while the list of topics focused on in the Samyutta seems to adumbrate the topics that are so prominent in certain of the canonical Abhidhamma works. A comparison of the Pali Saṁyutta Nikāya with what we know of other Saṁyukta recensions shows that in essence the Saṁyutta/Saṁyukta method consists of compiling and working up a body of sutta material around the following lists: (1) the 5 aggregates, (2) the 6 sense -bases, (3) the 12 links of dependent arising, (4) the 4 focuses of mindfulness, (5) the 4 right strivings, (6) the 4 paths of success (7) the 5 faculties, (8) the 5 powers, (9) the 7 awakening-factors, (10) the noble eightfold path. 341 In fact, these 10 lists appear to constitute a consistent core element of the Saṁyutta/Saṁyukta collections, attracting the most attention in the Pali version and it seems in the recension surviving in Chinese translation. Further, a number of other lists seem to act as important satellites, especially the 4 noble truths and the 4 dhyanas. This core list of lists continues to be of great importance in the later history of Buddhist thought and literature. 342 We find it expanded and developed as the basis of such canonical Abhidhamma/Abhidharma works as the Vibhaṅga, Dhātu,kathā, and Dharma,skandha, 343 and also such later works as the Arthavi- niścayasūtra.

5.3.3.4 Gethin discusses problems we are likely to face when we try to trace the development of this core mātikā.344 For example, in the canonical texts, the 4 truths and the 4 dhyanas find a firm place in the core, while the most consistent additions common to all versions appear to be the 5 precepts and the 4 immeasurables, neither of which feature at all in the Pali or Chinese Saṁyutta/Saṁyukta collections. Of course, focusing on this core mātikā in this way tends to the view, as expressed by A K Warder, that the earliest Abhidhamma/Abhidharma simply consisted of this mātikā/mātṛkā, and that it is the Vibhaṅga, in the case of the Pali Abhidhamma, that represents the earliest and basic Abhidhamma text. According to Warder, the Dhamma,saṅgaṇī, with its elaborate mātikā of triplets and couplets represents a somewhat later refinement.

5.3.3.5 However, in an important but neglected section of the introduction to his edition of the Abhi- dharmadipa,345 P S Jaini presents a considerable body of material, the effect of which is to call into question the adequacy of such a view of the development of the early Abhidhamma. Jaini himself expresses certain doubts in his review of Warder’s essay but does not pursue the matter. 346

341 Cf M Anesaki, “The four Buddhist Agamas in Chinese” 1908: 1-149 (esp 68-126); Bronkhorst, “Dharma and Abhidharma,” 1985:316 f. 342 Cf Warder, “The Mātikā,” 1961:xx. 343 For the Dharmaskandha, see J Takakusu, Journal of the Pali Text Society 1905:111-115. 344 See eg Warder, “The Mātikā,” 1961; Bronkhorst, “Dharma and Abhidharma,” 1985. 345 Abhidharmadīpa with Vibhāṣāprabhāvṛtti, ed P S Jaini 1959:22-49 (esp 40-45). 346 BSOAS 26, 1963:438-39.

72 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

It is, however, worth reflecting on the place of the triplet-couplet mātikā a little further. What Jaini points out is that the triplet-couplet system of analysis is not peculiar to the Pali Abhidhamma method, but, on the contrary, is also fundamental to the dharma analysis of works such as Vasubandhu’s Abhidhar- ma,kośa, Asanga’s Abhidharma,samuccaya, and the Abhidharma,dīpa itself, except that in these works, the number of triplets and couplets employed is somewhat reduced. 347

5.3.3.6 In fact, as a supplement to Jaini’s findings [8.3.3.5], it is worth noting, says Gethin, that the gap between the number of triplets and couplets distinguished in the Dhamma,saṅgaṇī and in the northern Abhidharma sources perhaps appears greater than it really is.348 Further, certain triplets already are found in the earlier sutta sections of the Pali canon.349 Thus it would seem that the kernel of the triplet/- couplet mātikā may be very ancient, and to regard either the core mātikā beginning with the 5 aggregates or the triplet/couplet mātikā as more fundamental than the other is to misunderstand the basic principle that determines the way in which the Abhidhamma develops out of the use of mātikās. When we examine the Dhamma,saṅgaṇī and the Vibhaṅga, we will see the triplet-couplet mātikā and the core mātikā as acting like the 2 axes of the Abhidhamma method. The Dhamma,saṅgaṇī treats the core mātikā by way of the triplet-couplet mātikā, and the Vibhaṅga treats the triplet-couplet mātikā by way of the core mātikā . The important point, however, is that the 2 mātikās are fundamental to both texts. Indeed, Gethin suggests, the Abhidhamma method consists precisely in the interaction of the two mātikās, and that the Abhidhamma system is actually born of their marriage. Certainly, one of the characteristics of the use of mātikās in the Abhidhamma is the treatment of one list of categories by the categories of another list. Thus, the two lists act like the two axes of a graph table. This is precisely why Abhi- dhamma material is so susceptible to presentation by charts.350

5.3.3.7 We can see here how the mātikā or matrices were used by the early matrix-experts (mātikā,- dhara) to form helpful lists from the cornucopia of the Buddha’s teachings, like enjoying a buffet from sumptuous and varied meal-offerings. Such lists were mnemonic summaries for us to remember teach-

347 Abhidharma,kośa and Abhidharma,dīpa (written in response to it) both use 5 triplets and 15 couplets, while the Abhidharma,samuccaya uses 6 triplets and 22 couplets; most of these triplets and couplets have their counterparts in the mātikā of the Dhamma,saṅgaṇī, but not all of them. 348 The Dhamma,saṅgaṇī gains 4 of its triplets by simply taking 4 existing triplets and introducing a secondary principle (ie, the notion of “object” (ārammaṇa) of consciousness; thus triplets 9, 13, 19, 21 are variations on 8, 12, 18, 20, respectively). With the couplets the number is brought up to 100 by applying what are more or less the same 6 principles to 10 different lists of unwholesome categories. Of course, since the lists are different, when the resulting couplets are applied, say, in Vibhaṅga's “question” sections, this can result in significant differences in the answers. Nevertheless, there is considerable overlap here, and in the case of the “knots” (gantha), “floods” (ogha), and “bonds” (yoga) there is simple repetition (see Dhs 24). One suspects that the purpose in part was simply to reach the number 100. Finally, one should also perhaps bear in mind that the northern sources in question are later summary Abhidharma manuals that may have pared down the number of triplets and couplets to essentials; the triplets and couplets are not treated fully in Visuddhi,magga, a comparable Pali summary work. (Gethin 1994b n64) 349 Seven are found in Saṅgiti S and Das’uttara S: triplet 2 (D 3:216,19-20, 275,1-3); triplet 6 (D 3:217,1-2, 274,25- 28); triplet 11 (D 3:218,1 f, 219,3 f); triplet 14 (D 3:215,23 f); triplet 15 (D 3:217,1 f); triplet 18 (D 3:216,16 f); triplet 22 (D 3:217,22-34). The “within/without/within-without” triplet has an important place in (Mahā) Satipaṭṭhāna Sutta; in the Aṅguttara, we have what appears to be the “small/become great/immeasurable” triplet (A 5:63). Thus, a total of 9 triplets have explicit Nikāya antecedents. Curiously, the triplet that appears to be most basic in both the southern and northern systems, the “wholesome/unwholesome/indeterminate” triplet, is apparently absent from the Nikāyas, but it is found in the Vinaya, though not in the oldest portions (V 2:91 f). 350 Cf Norman, Pali Literature 1983:107 http://dharmafarer.org 73 SD 58.1 The Oral Tradition of the Early Buddhists ings, such as the 7 sets, any of which can be used for our practice and for understanding the fruits of our practice. This is a vital feature of the early Buddhist oral tradition. In due course, this habit of listing teachings for practice to free the mind, evolved amongst the technically-inclined monastics into the practice of listing teachings to fix the teachings with the hope of some of philosophical accuracy. Such lists go beyond listing teachings into classifying mental states, and this can be such a fascinating exercise. But the Buddha is not a butterfly collector; he is a keen observer of nature, watching it “changing while it stands.” However, when the Abhidhamma brings us back to inner seeing, then we may rely on it as a Dharma-spirited guide.351 We must now resume our quest for some useful understanding of the early Buddhist oral tradition.

5.4 AṄGA: LIMBS OF THE TEACHER’S TEACHING

5.4.1 The limbs of the teaching

5.4.1.1 We must imagine the newly awakened Buddha sitting alone under the Bodhi tree. In due course, he teaches the 5 monks, and then the 60 monks [1.1.1.2]. Clearly, the first 60 disciples were spiritually intelligent and insightful, right and ready to hear the Buddha’s teaching and transform themselves into arhats. They were by nature spiritually inclined to a life of renunciation, and by nurture, they were naturally insightful, on account of their present practice, such as living as ascetics even before meeting the Buddha. For such special individuals to meet the Buddha and hear him teaching, they all responded positively like fish in a drying pond finding an opening into a large flowing river. Like well-trained specialist scientists listening to a new radical theory of everything, they easily understand what the scientific genius is talking about. The Buddha simply speaks his mind and heart; the first disciples heard and felt the teaching and awakened to the truth. The first 60 converts, and those who immediately follow after them, are not merely monastics: they are Buddha-like in every way, except that the Buddha arises well before any of them [5.2.2.2]. Looking at any of these arhats is just like looking at the Buddha himself [1.2.2]. For anyone who is open-minded, especially when he is wise, just a word from the Buddha, just a Buddha-word, is sufficient. He is at once connected with the Dharma, and often, he renounces the world. When they are moved by the Dharma, renounce, but remain unawakened, then they need to move further in the Dharma. On their account, the Buddha and his arhats give a liberating range of teachings in a direct and personal way—an oral transmission—that, in time, awakens them (and us) to the path of liberation. It is this oral tradition that we must seek and master. Understandably, these teachings begin simply as untitled profound personal conversations that open our mind and free our heart. They are called suttas, “threads” that go back to the Buddha himself, transmitting the Buddha-word, but they have neither name nor title at first. They were like songs happily and sweetly sung impromptu to music that arises spontaneously, sung with radiant breath, free from bodily limits. As this music and magic spread, names are given to them as more are touched and turned by them.

5.4.1.2 With awakening, there is silence, a truly beautiful stillness; the unawakened could not hear it. By the Buddha, silence becomes word; the word opens the mind, stills the heart. It is a nameless yet profound silence, a silence that makes all sounds possible and meaningful; the silence that frees. Silence is the Buddha’s wisdom, sound his compassion. Named sounds become words so that we know where to look, how to listen to the silence once again. These are the suttas.

351 See Sujato’s comment on Abhidhamma: SD 13.1 (3.9y.3.6).

74 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

In the Pali canon itself, the Buddha’s teachings, as a collection, is most often called dhamma,vinaya (the teaching and discipline) throughout the 4 Nikāyas;352 and also buddha,vacana (the Buddha-word),353 or pāvacana (the word).354 The suttas is also simply called the dhamma,”teaching,” and listed out in 9 genres or literary categories,355 known in the Commentaries356 as the 9 limbs of the Teacher’s teaching (nav’aṅga satthu,sāsana).

5.4.1.3 Before discussing the 9 limbs or aṅgas [5.4.2], we need to examine the existence of shorter lists of aṅgas in the suttas, and the likelihood that the list of 9 (and 12)357 had evolved from them. We may imagine that before the 9 aṅgas, at an early stage, there was some shorter lists which, in time, we found to be unfeasible, and were replaced by the list of 9. The fact remains that the aṅga lists existed in different sizes, and each suits the teaching of a particular sutta. We see shorter listings of aṅgas in at least 2 suttas. The first is found in the Kāraṇa,pālī Sutta (A 5.- 194), which relates how the brahmin Kāraṇa,pālī declares that anyone who has heard the Buddha’s teaching in the form of sutta, geyya, veyyākaraṇa, and abbhuta,dhamma will no longer be interested in the teachings of others.358 We can but guess why Kāraṇa,pālī mentions only 4 of the 9 aṅgas. But then, we are more likely to be surprised than impressed should he, as a brahmin—who is probably only familiar with the Buddha by reputation—have known all the 9 aṅgas at this point. He is probably listing the aṅgas as he recalls them impromptu. This also suggests that the aṅgas may have always been a list of 9, and that even so, at times, only the key ones are mentioned in certain suttas according to context. This can only be conclusive after an exhaustive survey of the occurrences of the aṅgas in the suttas. As CHOONG Mun-keat has suggested, this shorter list “would then simply be a case of abbreviation.”359

352 Dhamma,vinaya: V 3:90; D 1:8, 3:117; M 1:284, 457, 2:243; S 1:9, 3:12, 5:144; A 1:36, 121, 3:138, 229; U 50 353 Buddha,vacana: V 1:40, 2:17x2, 139x5, 4:54; Sn 202; S 2:259, 261, 262, 4:283, 292, 297; Tha 403; Ap 6.97/1:44 nav’aṅga,buddha,vacanaṁ (but it also mention “Abhidhamma” as a collection). 354 Pāvacana: V 3:107x2, 108; D 1:88, 2:154, 3:195, 206; M 1:488, 2:169; Tha 587; Thī 457, 477. 355 For the list of 9 genres: A 2:7x4, 4.102/103x4, 108x3, 108, 178, 185x4 186x4, 187, 3:86, 87, 88x2, 177x2, 237, 362, 362, 4:113; M 22,10/1:133, 134x3, 3:115; V 3:8, 9. Interestingly, this list of 9 is mentioned in neither Dīgha (esp D 33 + 34) nor Saṁyutta. See Ency Bsm 1:616-619 sv aṅga. 356 Buddhaghosa discusses the 9 aṅgas in Samanta,pāsādikā (VA 1:28 f), Sumaṅgalā,vilāsinī (DA 1:23 f) and Attha,- sālini (DhsA 1:26 66). However, many of his views seem to reflect his times rather than these aṅgas as they were understood in the Buddha’s time or in the suttas. 357 Mahāyāna has the “12 limbs” (Skt dvādaśâṅga)—also called vacana or pravacana—comprising the 9 traditional limbs—sūtra, geya, vyākaraṇa, gāthā, udāna, itivṛttaka, jātaka, adbhutadharma, vaipulya, plus: nidāna, stories (framing accounts); avadāna (heroic legends) and upadeśa (instructions). 358 A 5.194,1.5(1)/3:237,17+23 (SD 45.11). 359 Qu in Analayo (2016:22 n46): In reply to the hypothesis by von Hinüber 1994 that this passage points to an early stage in the evolution of the aṅgas, Choong argues that it is “likely that the unique Pāli list of just four aṅgas … is, rather, an abbreviation of the entire set of nine aṅgas in their original sequence; that is ‘sutta, geyya, veyyā- karaṇa, … abbhuta,dhamma’.” (2010:60). Analayo agrees that “the idea that these 4 aṅgas were an early division of the textual material is also not easily reconciled [with his discussion in 2016:21 f], in that a reciter who specializes on marvels would have relatively little material to learn and would moreover stand good chances to acquire an un- balanced understanding of the teachings.” (id). http://dharmafarer.org 75 SD 58.1 The Oral Tradition of the Early Buddhists

5.4.1.4 The other Sutta giving a shorter list of aṅgas mentioned above [5.4.1.3] is the Mahā Suññata Sutta (M 122). which mentions only sutta, geyya and veyyākaraṇa.360 The same triad is found in a Chinese parallel in the Madhyama Āgama, whereas a Tibetan parallel gives a full set of 12 aṅgas.361 In the context of the Mahā Suññata Sutta and its parallels, the Buddha declares that we should not follow a teacher merely for the sake of sutta, geyya, and veyyākaraṇa, but for the sake of beneficial teachings. The Sutta and its parallels agree that such beneficial talks are on the topics of the full path of liberation, that is, (in essence) on morality, concentration, wisdom, freedom (awakening), and the knowledge and vision of freedom. 362 In view of this narrative context, the earlier reference to sutta, geyya, and veyyākaraṇa (or the 12 aṅgas) could not be to a scheme representing the whole of the teachings of the Buddha and his disciples.363 Logically, if all the textual collections comprising the teachings given by the Buddha and his disciples are set apart as insufficient grounds for following him, there would be nothing left for the sake of following him! (Analayo 2006:23). The Buddha’s laying out of the path of rightness in its practical components, in fact, encompasses the whole of the teaching. Hence, the import of the Buddha’s remark is that we should not mistake what are merely aspects of the teaching for the whole teaching. We can reasonably conclude, then, that the Sutta passage is not about a listing of textual divisions representing all of the teachings or transmitted texts. What is it about then? [5.4.1.5]

5.4.1.5 Clearly, then, Mahā Suññata Sutta (M 122) [5.4.1.4], which is an early Buddhist text, in this important passage, is stating that we should not follow a teacher merely for the sake of explanations (vey- yākaraṇassa hetu) of teachings (sutta) or recitations (geyya)—suttaṁ geyyaṁ veyyākaraṇassa—for ac- cumulating knowledge, but rather for putting what we have learned (through listening and reciting) into practice.364 We should then conclude that the reference to the 3 aṅgas in the Mahā Suññata Sutta and its Ma- dhyama Āgama parallel as simply referring to the texts transmitted orally by the Buddha and his disciples. In other words, the reference to 3 aṅgas does not reflect an early stage in the evolution of the aṅgas, nor should we dismiss it as being the result of a textual corruption. Rather, it is what it says: we should not merely take the Dharma merely as expositions on learning and recitation, but for the sake of its practice leading to awakening.

5.4.1.6 Japanese scholars have proposed that the listing of the 5 aṅgas formed the origin for the 9 or 12 aṅgas.365 The Sanskrit *Karmavibhaṅgopadeśa gives a list of 6 aṅgas.366 According to Analayo, this actually seems to be the result of a textual corruption, as the same *Karmavibhaṅgopadeśa continues right away by summing up its presentation to be about the 9 aṅgas.367

360 M 122,20.1/3:115,18 (SD 11.4): see n ad loc. For a critical reply to the suggestion by Sujato that Skt fragments of the Mahāparinirvā-sūtra support the notion of a special emphasis being accorded to the first 3 aṅgas 2005:62; cf Anālayo 2011a:698 n69. 361 MĀ 191 (T1.739c4) and Skilling 1994: 242,13. 362 M 122,20.2/3:115,25; MĀ 191 (T1.739c8), and Skilling 1994:244,13. 363 Analayo 2016:23 n50: pace Choong 2000:9 f and Sujato 2005:61 f. 364 This is, in fact, proposed by Comy on the Sutta (MA 4:164,9), which says that, even though gaining great learning has been compared by the Buddha to a soldier gaining weaponry, the learning gained does not function just as weaponry that is not put into practice (Ñāṇamoli, The Greater Discourse On Voidness, 1982:30). 365 Mayeda, Genshi Bukkyō seiten no seiritsushi kenkū, 1964:25, 34; Nakamura, Indian Buddhism, 1980:28. 366 Another reference to 3 aṅgas can be found in Nett 78,9, comprising sutta, veyyākaraṇa and gāthā. 367 Analayo 2016:25. Lévi, Mahākarmavibhaṅga: sūtraṁ geyaṁ vyākaraṇam itivṛttam gāthodānam, evaṁ navāṅ- gaśāsanaṁ (1932a: 161,8). Kudo, The Karmavibhaṅgopadeśa, reads evan for evaṁ (2012:106).

76 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

It seems fair to conclude, then, that sutta references lists of aṅgas less than 9 aṅgas are not evidence of any early stage in the evolution of the 9 or 12 aṅgas.368 Rather, such references should be carefully examined in their respective sutta contexts (and the Commentaries) to let the suttas speak for themselves. This simple gesture often brings rewarding surprises in the study of the early Buddhist texts that have been carefully coded for our benefit even today.369

5.4.2 Nav’aṅga satthu,sāsana: the 9 limbs of the teacher’s teaching

5.4.2.0 The 9 limbs of the teacher’s teaching (nav’aṅga satthu,sāsana) are as follows:370

(1) sutta the suttas or discourses, ie, prose passages, such as those in Sutta Nipāta, Nid- desa, Vinaya, and texts with “Sutta” in their titles; [5.4.2.1] (2) geyya the mixed prose and verse (sa,gāthaka), such as the Sagāthā Vagga of Saṁyut- ta (S 1), Kasi Bhāra,dvāja Sutta (Sn 1.4/12-26); [5.4.2.2] (3) veyyākaraṇa the expositions, that is, elaboration of brief teachings of the Buddha; [5.4.2.3] (4) gāthā the verses, eg Dhammapada, Thera,gāthā, Therī,gāthā, Sutta Nipāta verses; [5.4.2.4] (5) udāna the inspired utterances, especially the Udāna and those in the suttas; [5.4.2.5] (6) iti,vuttaka the sayings, such as the Iti,vuttaka; [5.4.2.6] (7) jātaka the birth stories, such as those in the suttas, and the Jātaka verses (the 10th book of the Khuddaka Nikāya); [5.4.2.7] (8) abbhuta,dhamma the marvels, special qualities of disciples; [5.4.2.8] and (9) vedalla the answers to questions (catechetic suttas).371 [5.4.2.9]

Interestingly, this set of 9 limbs is neither mentioned in the Dīgha Nikāya—neither in the Saṅgīti Sutta (D 33) nor in the Das’uttara Sutta (D 34)—nor in the Saṁyutta Nikāya. We may surmise its absence from the Dīgha as suggesting that it is not a teaching for non-Buddhists [3.2.2.1]. It is most often found in the Aṅguttara probably because it is meant for teaching the laity or novices: as we shall see, they are only a broad guide to the kinds of teachings available to the teacher. It is rarely mentioned in the Vinaya and the Majjhima: the set is probably not used in the teaching of the more seasoned monks. Note that in the Alaggadûpama Sutta (M 22), the set of 9 limbs of the teaching is mentioned to the recalcitrant monk, Ariṭṭha, reminding him of the true teaching and its real purpose. [5.4.9.11]

368 Variations is also seen in similar lists in the Jain texts, eg, Viyāhapaṇṇati refers to 11 aṅgas only [Lal- wani, Sudharma Svāmī’s Bhagavatī Sūtra, 1973:177,14], differing from the standard count of 12 that forms the foundation for Śvetāmbara canon. For a survey of the 12 aṅgas of Śvetāmbara Jain tradition, see Dundas, The Jains, 1992:64 f. 369 Hinüber’s hypothesis that the listing of 4 aṅgas reflects an early attempt at organizing the texts (1994), is rejected by Klaus, that such a view is not supported by the texts (suttas), which do not present the aṅga lists as an attempt at categorizing the different texts, but rather as attempts at classifying or simply enumerating them, “mir kommt es auf die Feststellung an, daß Vermutungen in diese Richtung sich nicht an die Texte anknüpfen lag themss- en. Die Texte präsentieren uns die verschiedenen Aṅga-Listen nicht als Versuche, einen wie auch immer gearteten Gesamtbestand an Texten zu ordnen, sondern als Versuche, die verschiedenen Arten von Dhamma-Texten zu klass- ifizieren oder auch nur aufzuzählen” (2010:518). Cousins succinctly says that “short versions are sometimes interpreted as earlier lists of ‘Aṅgas,’ but that seems quite anachronistic to me.” (2013:105) 370 Explained at MA 2:106,8-28 = VA 28,8-29,3. For refs to the term: DA 1:23; MA 1:133 f, 5:109; UA 4; ItA 1:2; VvA 4; PvA 2; ThaA 1:2; ApA 103; CA 2; NmA 1:10; PmA 1:4, 9; DhsA 26; Miln 263. For nn: SD 51.16 (2) defs; SD 3.2 (1.4); SD 26.11 (3.2.1.3); SD 30.2 (2.1); SD 30.10 (4). 371 On these 9 limbs (aṅga), see Mahā Vedalla S (M 43), SD 30.2 (2.1). http://dharmafarer.org 77 SD 58.1 The Oral Tradition of the Early Buddhists

5.4.2.1 Sutta (1) Significantly, sutta is the 1st of the 9 limbs or aṅga. Although it is only one of these “9 limbs of the Teacher’s teachings,” sutta often generically refers to any or the whole of the ninefold set,372 that is, underpin them as the Dharma as teaching, as contrasted from the Vinaya, the monastic discipline and texts, and from the Abhidhamma, the later scholastic developments. In other words, sutta refers to teachings given by the Buddha or his immediate disciples that has been “threaded” (sutta) together in the Sutta Piṭaka or, where a teaching does not go against any the spirit of the Buddha Dharma.373 The Vinaya records the Buddha as recounting how past buddhas like Kakusandha, Koṇāgamana and Kassapa were diligent in teaching the Dharma “in detail,” that is, by way of the 9 limbs mentioned, to their disciples, along with formulating training-rules and instituting the Pātimokkha, their dispensation lasted long. He then uses the analogy of “string together” flowers to form a garland, which otherwise would be scattered by the winds.374

(2) Sometimes the term sutt’anta is used for “the teaching” as sutta is used, although it is sometimes used only in the titles of longer discourses, especially those of the Dīgha Nikāya. In the term, dhamma,- vinaya, dhamma refers to suttas, but can sometimes include the older sections of the Abhidhamma.375 The term sutta is also often found in the Vinaya, where it usually refers to the individual rules, and to a lesser extent in the Abhidhamma (such as the Vibhaṅga and the Puggala Paññatti, the 2nd and the 4th book of the Abhidhamma, respectively).376

(3) Hence, Buddhaghosa says: “The twofold Vibhaṅga, the Niddesas, the Khandhaka, the Parivāra, Maṅgala, Ratana, Nālaka, and Tuvaṭaka Suttas of the Sutta Nipāta377 and other sayings of the Tathāgata bearing the name sutta should be known as Sutta (Discourses).” (MA 2:106,8 = VA 28,8-11)378 The 2 Vibhaṅgas are those dealing with the monks’ rules (bhikkhu,vibhaṅga, V 3) and the nuns’ rules (bhikkhuṇi,vibhaṅga, V 4), that is, the Mahā,vibhaṅga and the Bhikkhuṇī Vibhaṅga respectively, constituting the Sutta Vibhaṅga of the Vinaya.379 The Niddesa, the 11th book of the Khuddaka Nikāya, comprising the Mahā,niddesa (Nm) and the Cūḷa,niddesa (Nc), are canonical commentaries to the oldest parts of the Sutta Nipāta [4.1.1.2]. The Khandhaka, comprising the Mahā,vagga (Mv = V 1) and the Culla,vagga (Cv = V 2), is based on the monastic rules, set in the framework of the Buddha’s life [Norman 1983b n below]. The Parivāra (V 5) is a supplement to the Vinaya, forming its last book, and of uncertain but no earlier than 1st century BCE.

372 Technically, we also have the categories “Vinaya” (referring to the training rules and their background) and “Abhidhamma” (the later philosophical and psychological lists and teachings). 373 According to Przyluski, the use of the expression sutta in the context of the aṅgas has the specific sense of a discourse that begins with a numerical exposition of a teaching, “un sūtra était un sermon commençant par un ex- posé numérique” (eg, “there are 4 things … what are the four,” etc) (Le concile de Rājagṛha 1926:341). Ñāṇaponika explains that sutta is a presentation of a teaching that is, as it were, internally connected by a thread, “eine zusam- menhängende Lehrdarstellung … durch die sich ein gemeinsamer Faden hindurchzieht.” (Sutta-Nipāta 1977:13 f). 374 Pār 1.3.3 (V 3:8 f) 375 See eg, Dhamma,vihārī S (A 5.74,2/3:88), SD 44.5; also SD 30.10 (4); SD 26.11 (3.2.1.3). 376 For publications related to sutta that derives from su + ukta, “well spoken,” see Anālayo, A Comparative Study of the Majjhima-nikāya, 2011a:150 n22. 377 These are respectively: Sn 16 (SD 101.5), 13 (SD 101.2), 37 (SD 49.18) and 52 (SD 101.6). 378 See Jayawickrama (tr) Inception of the Discipline, 1962:25 f. 379 On Vinaya, see Norman 1983b:18-29: 2.1 Sutta,vibhaṅga, 2.2 Khandhaka, 2.3 Parivāra. For details of contents, see Shayne Clarke, in Brills’ Ency Bsm, BEB 1:61-87 Vinayas (esp 60-63).

78 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

5.4.2.2 Geyya Geyya refers to texts “with verses” (sa,gāthakaṁ, VA 1:28,11). In others, they may be entirely of verses, or may contain a significant amount of verses. Both Buddhaghosa and the Sanskrit sources seem to agree on this. Hence, Buddhaghosa designates the Sa,gāthā Vagga (S 1/1:1-240), the opening book of the Saṁyutta, as geyya (DhsA 26,14 f).380 The literary limb seems to overlap with “the verses” or gāthā [5.4.2.4], the only difference being that geyya has prose in it, too. Apparently, in the case of geyya, the prose not only gives some narrative background but is a vital feature serving either as its narrative background or some kind of commentary to the verse portion. Both the Udāna [5.4.2.5] and the Iti,vuttaka [5.2.4.6] comprise prose and verse, and, as such, are also geyya. However, in either case, the prose are incidental stock passages that simply keep each sutta intact in itself. The It prose, however, often gives the background of the verses which highlight the teaching. From this and similar features in other limbs, we get a hint of the nature of the aṅga system, then, that it is not a listing of literary genres, but rather a description of how sutta materials are used as practical media of Dharma instruction and inspiration.

5.4.2.3 Veyyākaraṇa or vyākaraṇa (1) As a limb of the teaching, veyyākaraṇa is exemplified by the 10 suttas of the Vibhaṅga Vagga of the Majjhima Nikāya that preserve teachings by the early arhats, especially Mahā Kaccāna, elaborating on brief statements (saṅkhittena desitaṁ) made by the Buddha.381 The Abhidharma,samuccaya, a Sanskrit work attributed to Asaṅga (4th century CE), similarly says that veyyākaraṇa includes the expositions on the sutras as given by the early disciples (Abhsm 78). Buddhaghosa views that the Abhidhamma Piṭaka as a whole comes under this category.382 However, this is an anachronism, since the Abhidhamma as a Piṭaka (“basket” or collection) arose only after the Buddha. He also says that “any sutta without verses” (niggāthakaṁ suttaṁ) should be regarded as veyyā- karaṇa (VA 1:28,13).383

(2) Buddhaghosa’s suggestion that veyyākaraṇa refers to suttas without verses is not born out by those suttas that use the term to refer to themselves, since several of these do have verses. The Sakka,- pañha Sutta (D 21), for example, has these long verse passages: §1.5/2:265 14 quatrains; §1.7/2:268 14 quatrains; §1.12/2:272-275 19 quatrains + the Buddha’s verse 20 quatrains;

380 Jayawickrama says that “geyya (from √GAI, gāyati, to sing), seems to represent the ākhyāna-type containing stanzas punctuated with narrative prose.” (“Buddhaghosa and the traditional classification of the Pāli canon,” UCR 1959:12). According to Mayeda, geyya “is not, however, a simple juxtaposition of prose and verse. The prose section which comes first is repeated once again in the verse section which follows. This repetition of similar contents is the key point of geyya.” (“A history of the formation of early Buddhist texts,” 1964:24). Cf Burnouf, Introduction a l’histoire du buddhisme indien 1844:47. 381 These 10 suttas of Vibhaṅga Vg (M 131-142) of Majjhima are: Bhadd’eka,ratta S (M 131/3:187-189), Ānanda Bhadd’eka.ratta S (M 132/3:189-191), Mahā Kaccāna Bhadd’eka.ratta S (M 133/3:192-199), Lomasak’aṅgiya Bhad- d’eka,ratta S (M 134/3:199-203, taught by the Buddha himself), Cūḷa Kamma Vibhaṅga S (M 135/3:202-206), Mahā Kamma Vibhaṅga S (M 136/3:207-215), Saḷāyatana Vibhaṅga S (M 137/3:215-222), Uddesa Vibhaṅga S (M 138/- 3:223-229), Araṇa,vibhaṅga S (M 139/3:230-237), Dhātu Vibhaṅga S (M 140/3:237-247), Sacca Vibhaṅga S (M 141/- 3:248-252), Dakkhiṇa Vibhaṅga S (M 142/3:253-257). 382 VA 1:28,13-15. 383 For a detailed overview of veyyākaraṇa, see Anālayo, Ency Bsm 8 2008b:549-554. The foll section is from this source with additional comments. http://dharmafarer.org 79 SD 58.1 The Oral Tradition of the Early Buddhists

§2.8/2:285-287 6 quatrains; §2.9/2:288 f 9 quatrains; and concludes with the narrator saying: “Hence, this exposition is called the Sakka,pañha, ‘the questions of Sakra” (tasmā imassa veyyākaraṇassa ‘sakka,pañho’t’eva adhivacanaṁ, D 21,2.10.6/2:289).384 Another example is the Brahma Nimantaṇika Sutta (M 49), which contains verses (M 49,9/1:328, 27/1:330), and similarly concludes with “this exposition is entitled “Brahmā’s invitation” (imassa veyyā- karaṇassa brahma,nimantaṇikan’t’eva adhivacanaṁ, M 1:331).385 Thus, the commentarial explanation of veyyākaraṇa as suttas without verses does not fit the actual use of the term veyyākaraṇa in the Nikāyas.

(3) According to Asaṅga’s Śrāvakabhūmi (“stages of the disciple”), geyya differs from veyyākaraṇa in that while geyya stands for texts “that still need explanation” (neyy’attha), veyyākaraṇa refers to “what has already been explained” (nīt’attha).386 While this interesting explanation may work for some cases, it does not reflect the full range of meanings of veyyākaraṇa as found in the suttas. Veyyākkaraṇa may refer to an explanation “that has been drawn out,” but this depends not on its being an exposition (veyyā- karaṇa), but rather on the teaching being given in Dharma language.387 In the suttas, veyyākaraṇa does not always refer to an explanation that has been laid out. In the Susīma Sutta (S 12.70), for example, Susīma reports his puzzlement on hearing other monks making a statement or declaration (veyyākaraṇa) of final knowledge (S 12.70,23). The remainder of the Sutta is devoted to drawing out the implication of the veyyākaraṇa that Susīma has been unable to understand, so that at least in this case, the veyyākaraṇa made by the monks was, from Susīma’s perspective, clearly in need of further explanation (neyy’attha).388

(4) Another example is the Kīṭā,giri Sutta (M 70), which uses veyyākaraṇa to refer to a succinct saying, the “4-line exposition” (catu-p,pada veyyākaraṇa, M 70,25). The situation here is interesting in a context- ual sense. According to the Sutta, such a saying would be explicit, nīt’attha, to “the wise … (who) would quickly understand it through wisdom,” but would remain implicit, neyy’attha, for others.389 The term veyyākaraṇa introduces the Sutta’s closing verse, given in reply to the key question in the closing story told by the Buddha in the Kevaddha Sutta (D 11/1:223). It has the same open senses as in the Kīṭā,giri Sutta. Here, however, it should be clear, that is, “explicit” (nīt’attha) to the questioning monk.390

(5) Veyyākaraṇa is not only part of the standard listing of 9 aṅgas, but also occurs in abridged listings of the aṅgas. Analayo (2008b:551] cites the case for “a threefold listing” found in the Mahā Suññata Sutta (M 122), which reads sutṭaṁ geyyaṁ veyyākaraṇassa hetu, which should be translated as “for the sake of the exposition of sutras [formulas] or of geyas [recitations].”391 Hence, actually, only 2 limbs are mentioned here. This also attests to the polysemy of veyyākaraṇa, whose sense has to be deduced out from its context.

384 D 21/2:263-289 (SD 54.8). 385 M 49/1:326-331 (SD 11.7). 386 On Śrāvakabhūmi, see Yasuo Matsunami, Śrāvakabhūmi: yac ca sūtraṁ neyārthaṁ, idam ucyate geyam ... yac ca sūtraṁ nītārtham, idam ucya te vyākaraṇam (1998:228); the same work also takes geya to stand for text with verses and vyākaraṇa for proclamations about the rebirth of disciples. Cf Ruegg: “Remarks on the place of narrative in the Buddhist literatures of India and Tibet,” 1999:206. 387 On these 2 terms, Neyy’attha Nīt’attha S (A 2.3.5+6/1:60), SD 2.6b; SD 47.20 (1.3); SD 89.10 (1). 388 S 12.70,23/2:123 (SD 16.8). 389 M 70,25/1:480 (SD 11.1). 390 D 11,85.4/1:223 (SD 1.7). 391 M 122/2:109-118 (SD 11.4).

80 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

The Chinese parallel also has only a triad of sutra geya vyākanaṇa (T1.739c4), while a Tibetan parallel lists altogether 12 aṅgas,392 a set often found in texts of the Mahāyāna texts.393 Another abridged listing occurs in the Kārana,pālī Sutta (A 5.194) from the Aṅguttara Nikāya, where the brahmin Piṅgiyāni declares he is deeply inspired by the Buddha’s teaching, “be it the suttas, or the mixed discourses, or the expositions, or the miraculous accounts” (vadi suttaso yadi geyyaso yadi veyyā- karaṇaso yadi abbhuta,dhammaso), thus comprising only 4 aṅgas.394 A sixfold listing can be found in the Sanskrit analysis of karma, Karmavibhaṅgopadeśa, that reads sūtraṁ geyaṁ vyākaraṇaṁ itivṛttaṁ gāthôdānam.395 At least in the last case, the shortened list is clearly the outcome of textual loss, as the same text continues to speak of the 9 angas, evaṁ nav’aṅga,śāsanaṁ.

(6) Thus, veyyākaraṇa or vyākaraṇa is clearly a key component of the aṅga listings. Yet, unless sutta originally had a meaning different from “discourse,”396 it would be difficult to understand why veyyākara- ṇa is mentioned separately, given that veyyākaraṇa regularly stands for a sutta. Moreover, in as much as some suttas that refer to themselves as veyyākaraṇa contain verses, it seems similarly difficult to draw a clear distinction in relation to geyya [5.4.2.2] (provided this term stands indeed representative for suttas that contain verses, as suggested by the Commentary (MA 2:106).

5.4.2.4 Gāthā Some of the oldest verse works have been preserved in the Sutta Nipāta (Sn), the 5th book of the Khuddaka Ni- kāya. Sn is, in fact, a collection of verse suttas in 5 chapters (vagga), with some prose insertions.397 The first 4 chapters—Uraga Vagga, Cūḷa,vagga, Mahā,vagga, and Aṭṭhaka,vagga—contains a total of 54 short poems; while the 5th chapter, Pārāyana,vagga, is a long independent poem, consisting of 16 shorter suttas, each named after its questioner (one of the “16 youths,” pupils of Bāvarī); hence, famously called the “16 questions” (solasa,pañha). Many of these ancient poems are quoted in other suttas.398 16 of the 54 suttas have prose sections, which technically would be regarded as geyya [5.4.2.2]. Buddhaghosa, quite justifiably, designates the Dhammapada, the Theragāthā and the Therīgāthā as belonging to this limb.399 However, he also includes here “sections entirely in verse in the Sutta Nipāta which are not called suttas (no sutta,nāmikā).” There seems to be no clear distinction between sutta and gāthā, except that prose is excluded from gāthā.400

392 Skilling, Mahāsūtras, 1994:242. 393 See survey in Lamotte, Le Traité de la Grande Vertu de Sagesse, 1980:2281-2287; Mayeda op cit 1964, for a comprehensive chart of aṅga listings; and the study by Nattier, "The twelve divisions of Scriptures in the earliest Chinese Buddhist translations," 2004:167-196. 394 A 5.194,1.5/3:237 (SD 45.11). 395 S Levi, Mahākarmavibhaṅga, 1932a:161. 396 Von Hinüber: “Die Neun Angas, Ein früher. Versuch zur Einteilung buddhistischer Texte,” 1994a:129 suggests that sutta could originally have intended just the pātimokkha-sutta, thence come to represent the Vinaya. The Comys indeed include the whole of the Vinaya under the heading of sutta (MA 2:106). 397 Sn suttas with some prose: Kasi Bhāra,dvāja S (Sn 1.4/12-16 ); Vasala S (Sn 1.7/21+24 f); Āḷavaka S (Sn 7.10/31 f); Sūci,loma S (Sn 8.5/47 f); Brāhmaṇa,dhammika S (Sn 2.7/50+54 f); Rāhula S (Sn 2.11/59); Vaṅgīsa S (Sn 2.12/59 f); Dhammika S (Sn 2.14/66); Subhāsita S (Sn 3.3/78); Sundarika Bhāra,dvāja S (Sn 3.4/79 f+86); Māgha S (Sn 3.5/86 f+91); Sn 3.6/91-103 passim); Sela S (Sn 3.7/102-108, 110 f, 111 f); Vāseṭṭha S )Sn 3.9/115 f, 123); Kokāliya S (Sn 3.10/123-127); Dvayatānupassanā S (Sn 3.12/139-149 passim); Pārāyana Vg ((Sn 5.18/218). 398 Cf R Otto Franke, “Die Suttanipāta-Gāthās mit ihren Parallelen,” ZDMG 63,1 1909a:1-64, esp 23 ff; 63,2 1909b: 255-286; 64,1 1910a: 1-57, 64,2 1910b:760-807; 66,4 1912:171-221; “Majjhimanikāya und Suttanipāta,” WZKM 26 1914:699-708. For a detailed survey poetry, esp narrative poems, see M Winternitrz, History of Indian Literature vol 2, 1933:76-165 Khuddaka Nikāya. 399 VA 1:28,15-17. 400 N A Jayawickrama, “Buddhaghosa and the traditional classification of the Pāli canon,” UCR 17,1 1959:13. http://dharmafarer.org 81 SD 58.1 The Oral Tradition of the Early Buddhists

5.4.2.5 Udāna The word udāna literally means “the breathing upwards.” Figuratively, it refers to an inspired or solemn utterance, mostly, but not necessarily, in verse form, and made without regard to any listeners.401 Although the 82 suttas of the Udāna (U), the 3rd book of the Khuddaka Nikāya, are given by Buddhaghosa as an example of this limb,402 this is clearly an anachronism. Even if the Udāna was assembled as a collection, surely this could not have been done during the 1st period of the Buddha’s ministry;403 perhaps, late in the 2nd period, but more likely during the 1st council. Clearly, the udāna as a limb of the teaching refers to teachings with inspired teachings scattered throughout the suttas.

5.4.2.6 Iti,vuttaka The compound, Iti,vuttaka, comes from iti, “thus” + vuttaka, “thus said”; hence, “the sayings (of the Buddha).”404 Like the Udāna, we have a collection called Iti,vuttaka (It), which is the 4th book of the Khud- daka Nikāya. Both collections end with verses, and both are always mentioned together in the various lists of aṅgas. The It suttas do not mention the Buddha’s location: they are all known to be spoken in Kosambī and heard and transmitted by the laywoman, Khujj’uttarā (Uttara the Hunchback).405 Unlike the Udāna, the prose and verses of It form a conceptual unit. In both case, U as well as It, are like the Sa,gāthā,vagga, and may be considered as geyya, too. Buddhaghosa is not wrong in quoting It as an example of the limb, iti,vuttaka,406 so long as we understand that it is not the only one, and that, like the Udāna [5.4.2.5], it was put together well after the aṅga system went out of vogue. Clearly, the 4th book of the Khuddaka Nikāya derived its name from the iti,vuttaka limb, as evident from the It’s opening words of most of its 112 very short suttas (mostly about a quarto page long): vuttam h’e- taṁ bhagavata, vuttaṁ arahatā’ti me sutaṁ, “This was indeed spoken by the Blessed One, heard by me spoken by the Arhat.”407 The term iti,vuttaka (as a limb) probably refers to the Buddha’s teachings with both prose and verse, like geyya [5.4.2.2]. However, unlike geyya (which may have longer instructive prose sections), the iti,vuttaka text tend to be very short teachings with very brief prose supporting or highlighting the verse section.

5.4.2.7 Jātaka On account of the huge number of Jātaka stories that existed in ancient India in the time of early Bud- dhism, it is clear that the Jātaka collection does not include all of them. Technically, a jataka is a story about the Bodhisattva (bodhisatta), the Buddha before his last life as Gotama. This is the “mythical” Bodhi sattva—the term is used in a psychological or archetypal sense, referring to one whom we today know will become the Buddha Gotama. The other, historical, sense of Bodhisattva refers to Gotama himself before

401 Mahv:G 130n; O Franke, D:F 207 n7. 402 VA 1:28,18 f. 403 On the periods in the Buddha’s ministry, see SD 1.1 (2.2); SD 40a.1 (1.3). 404 On the Skt title, see D S Ruegg 1999:201 f. 405 ItA 1:27,7-33,8; cf Hinuber 1996 §79. On Khujj’uttara, see SD 8.6 (8.2); SD 15.11 (1.3.3.1); SD 57.19b (1.2.2). 406 VA 1:28,20 f. 407 This phrase appears at the start; etam aṭṭhaṁ bhagavā avoca, tatth’etam iti (“The Blessed One spoke on this matter. In this connection, he said this …”) ends the prose sections and starts the closing verse section; and ayam pi attho vutto bhagavata iti me sutan’ti (“This matter, too, was spoken by the Blessed One. Thus I have heard”) closes every sutta—that is, until It 80/3.4.1. Thereafter, it appears only intermittently. The foll 27 omit those phrases: It 81- 88, 91-98, 101-111. Possibly, Catukka Vg (It 100-112) comes from Aṅguttara (it is not found in the Chin tr: K Wata- nabe 1906/7:44-49), but the rest are very ancient. That not all these It sayings are uttered by the Buddha may be deduced from the fact that the last sutta (It 112/4.13) refers to the Buddha in both the 3rd and 2nd persons. See Norman 1983b:62 f.

82 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists he awakens as the Buddha. Broadly, we can this this to be the meaning of jataka as a limb (aṅga) of the teaching. Buddhaghosa’s listing the Jātaka collection of 550 stories (VA 1:28,22 f) as an example of the jataka limb is an anachronism. The Jātaka stories consist of verses in a prose setting, of which, with very few exceptions, only the verses, called Jātaka,pāli, the 10th book of the Khuddaka Nikāya, are canonical. Then, there are accounts in the suttas which are technically jataka stories because they tell us of the Buddha in an earlier birth, and conclude with an identification of the characters, though these stories are not found in the collection proper. The best known of this is found in the Ghāṭīkāra Sutta (M 81).408 Further, we may include a number of canonical birth stories (in the sense of “stories of the past”), such as those found in the Kūṭa,danta Sutta (D 5), the Mahā Sudassana Sutta (D 17) and the Mahā,govinda Sutta (D 19).409 There are also several old Jātaka stories in the Mahāvastu, a Sanskrit work of the Lok’uttaravāda branch of the Mahāsaṅghika which have no Pali parallel.410

5.4.2.8 Abbhuta,dhamma The Attha,sālinī (Dhammasaṅgaṇī Commentary)411 defines abbhuta,dhamma as “All the suttas connected with wonderful and marvellous qualities spoken in this manner: ‘There are, bhikshus, 4 wonderful and marvellous qualities in Ānanda,’ are an account of the Buddha’s praising Ānanda’s special qualities in the Mahā,parinibbāna Sutta (D 16, SD 9), and its reprise in the (Catukka) Acchariya Sutta 3 (A 4.129).”412 Taking this definition as a cue, we may also include such suttas as the Sampasādanīya Sutta (D 28), SD 14.14, and the Lakkhaṇa Sutta (D 30), SD 36.9. On a more instructive level, we can include all suttas that centre on the lion-roar (sīha,nāda), that is, a declaration or act of great faith in the Dharma whether shown by the Buddha or his disciple, such as the following:

Udumbarikā Sīha,nāda Sutta D 25 the Buddha’s lion-roar: he shows how awakening may be attained only within 7 days, and teaches even when the audience fails to respond to his invitation. SD 1.4 (2.2) Bakkula Sutta M 124 Bakkula declares his self-reliance and solitary life throughout his 80 years of monkhood. SD 3.15 Piṇḍola Bhāra,dvāja Piṇḍola declares that anyone with doubt in the Dharma may approach him for clarification and succour; the foremost of monks who are lion-roarers. SD 27.6a esp (1.2) Dhātu Vibhaṅga Sutta M 140 Pukkusāti discovers his teacher through the Dharma (M 140,- 33 f), SD 4.17 Cūḷa Sīha,nāda Sutta M 11 the 4 noble individuals are not found outside of Buddha Dharma. SD 49.2 Kamma Mallikā Sutta A 4.197 Mallikā’s lion-roar about her own karma (A 4.197,7-18), SD SD 39.10 Nakula Sutta A 6.16,2 How Nakula,mātā counsels and comforts her gravely ill husband. SD 5.2

408 M 81 closes with the Buddha identifying himself with the brahmin youth, Jotipāla (ahaṁ tena samayena joti- pālo māṇavo ahosiṁ, M 2:54,18 f) in the story of the past (bhūta,pubbaṁ) he has related. 409 Respectively: D 5,10-20/1:134-143 (SD 22.8); D 17/2:169-199 (SD 36.12); D 19,29-61/2:230-251 (SD 63.4). 410 Eg, Puṇyavanta J (Mvst 3:33-41). For more details of jātaka, see Norman 1983b:77-84 (§3.5.10). 411 VA 1:28,23-26; DhsA 26,27-30. 412 This is an example of the Buddha’s referring to Ānanda’s charisma as a “marvel” (abbhuta,dhamma). Very likely, it is such passages (as D 16,5.15-16/2:144) that constitutes “marvels” (abbhuta,dhamma), rather than miraculous stories: see eg, Abbhuta,dhamma Ss (eg, Acchariya,abbhūta S, M 123) mentioned in the nn there. It is possible to incl lion-roars (sīha,nāda) here, too: see SD 36.10 (3). See also Ency Bsm: Aṅga (under abbhutadhamma). http://dharmafarer.org 83 SD 58.1 The Oral Tradition of the Early Buddhists

(Saddha) Subhūti Sutta A 11.15 The Buddha praises the monk Saddha as tathagata. SD 45.1 Cakka,vatti Sīha,nāda Sutta D 26 The wheel-turner’s lion-roar: the nature of Dharma-based power. SD 36.10 (3)

Of special mention as abbhuta,dhamma are accounts of miracles and wonders attending the life of the Bodhisattva and the Buddha, such as the following:

Acchariya,abbhuta Sutta M 123 Miracles and wonders attending the Bodhisattva’s birth and life. SD 52.2 (Catukka) Acchariya Sutta 2 A 4.128 4 marvels attending the Buddha’s arising. SD 31.6 (See also Acchariya,abbhuta Ss 1-4 (A 4.127-130). SD 52.2a+2b+2c+2d.

SD 52.1 is a special study volume dealing with the abbhuta,dhamma aspects of the Bodhisattva’s life. This is actually a study of early Buddhist mythology and an exploration into a psychology of mythology, that is, how stories, fables, legends and myths help us envision universal or special issues that depict the nature of being, our quest for humanity, divinity and awakening. Such stories use an “alternative language” of bizarre, poignant or “intentional”413 logic for us to envision meaning and purpose in life. In short, this is the stuff that defines literature and liberates the human spirit.

5.4.2.9 Vedalla Vedalla refers to suttas given in response to questions in the form of a catechism (repeated questions and answers) asked through knowing and joy (sabbe pi vedañ ca tutthiñ ca laddhā pucchita,suttantā). As examples, Buddhaghosa mentions the following: the Culla,vedalla (M 44), the Mahā,vedalla (M 43), the Sammā,diṭthi (M 9), the Sakka,pañha (D 21), the Saṅkhāra,bhājaniya [Saṅkhār’upapatti] (M 120) and the Mahā,puṇṇāma Suttas (M 109).414 Such texts, then, are “learned” discussions on the Dharma, often dealing with difficult, abstruse, even obscure, points. They are discussions between those who know and those who know better, such as between the Buddha (respondent) and a disciple (questioner), or between a wise elder and a junior elder.415

5.4.2.10 Evidently, the list of the 9 limbs of teaching is not an exact system of classifying teaching into neat categories. They must have been used at some point in early Buddhism to distinguish how the teachings can be used for specific purposes of learning or teaching the Dharma [5.4.2.2]. Further, as noted by Norman, “despite the fact that books called Jātaka, Udāna and Itivuttaka actually exist in Pāli, it is probable that the list of nine aṅgas did not originally refer to specific works in the canon, but was a description of various types of texts” (1983:16).416 Instead of being an early system for categorizing and transmitting the teachings that was eventually replaced by the division into āgamas or nikāyas, the aṅgas may have been just a listing of textual types. For such listings, the considerable overlap between individual aṅgas would not have been a problem, since the purpose of the listing would just have been to simply have a grasp of the whole variety of the early texts, like a guide to teaching or learning methods from which we can choose for our purposes. For such a purpose, the only point of importance would be to make sure that whatever is related is mentioned, for which the overlapping can even be helpful.

413 On “intentional” language, see SD 26.11 (6.5) ; Dh 97, SD 10.6 esp (5). 414 DA 1:24,15-18. The suttas are, respectively: M 44/1:299-305 (SD 40a.9); M 43/1:292-298 (SD 30.2); M 9/1:46- 55 (SD 11.14); D 21/2:263-289 (SD 54.8); M 120/3:99-103 (SD 3.4); M 109/3:15-20 (SD 17.11). 415 On the possible reason that vedalla is listed last, see Analayo 2016:18 f. On the possible original form and etymology of vedalla, see Karashima 2015:136. 416 For a view of the aṅgas as actual division of the canon, see Sujato, History of Mindfulness, 2005:48-65.

84 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

In contrast, the division into āgamas [5.4.3] or nikāyas [5.5] clearly reflects the exigencies of oral transmission, as it divides the body of texts into easily memorized portions. The aṅga list, then, has less to do with the sutta reciter (bhāṇaka) memorizing the texts [7.2], than with the Dharma teacher (dhamma,- kathika) given a versatile choice of teaching methods tools for his students. On a simpler level, he may simply use a sutta; for a more learned audience, he may choose a vedalla. In either case, he may still employ a veyyākaraṇa in the discourses, responses, declarations and expositions. Analayo’s conclusion on the nature of the aṅgas in his Encyclopaedia of Buddhism entry on “Veyyā- karaṇa” (Ency Bsm 8) is worth noting (and he is not alone in this): “Perhaps a solution to this conundrum can be found by seeing the nine aṅgas as overlapping categories. In fact, the difficulty of treating the aṅgas as neatly separate categories arises not only when trying to distinguish between sutta, geyya and veyyākaraṇa, but also in regard to the other members of the ninefold listing.417 Perhaps the listings of aṅgas, whether these comprise three or nine, was never intended to represent neatly separate categories. Instead, they may just have stood representative of the Bud- dha’s teaching as a whole.” (2009l:552).

5.4.2.11 The Buddha’s teaching as a whole, is famously said to be both its letter (vyañjana) and its spirit (attha), as mentioned in the pericope, “endowed with meaning and phrasing” (sâtthaṁ sa,vyañja- naṁ, D 1:62,31), that is, the Dharma is complete in both the spirit and the word (or letter), and should be when it is taught as the 9 aṅgas. The Dhammaññū Sutta (A 7.64) uses the 9 aṅgas to distinguish between mere knowledge (dhamma) of Dharma418 and understanding its meaning and purpose (attha). This distinction is echoed in Chinese parallels in the Madhyama Āgama, the Ekottarika Āgama and in an individual translation (though here in terms of the 12 aṅgas).419 According to this distinction, merely memorizing or knowing does not suffice for true knowledge, a theme also found in other references to the aṅgas. This vital significance of not merely knowing the teaching (dhamma), the text of the Dharma, but also its spirit (attha), that is, meaning and purpose, is often highlighted in other occurrences of the 9 aṅgas, as shown in the following suttas:

(Catukka) Appasuta Sutta A 4.6/2:7 f whether we learn little or much, what matters is that we do progress in Dharma; = Pug 4.28/62 f (without closing verses). SD 51.16 Valāhaka Sutta A 4.102/2:103 f we should learn and master Dharma for truly understanding the 4 truths, and inspire others by our Dharma spirit (not by only by our learning).420 Musika Sutta A 4.107/2:108 f a similar teaching as in preceding but with a different simile: one who masters the Dharma to truly understand the 4 truths is like a mouse who digs a hole and lives in it; = Pug 4.9/43 f. Dhamma,vihārī Sutta 1 A 5.73/3:86 f one who “lives the Dharma” (dhamma,vihārī), a true practitioner, understands it (by way of the 9 aṅgas) and diligently lives in seclusion

417 Jayawickrama comments that “there seems to be a good deal of overlapping, for the same piece can belong to several of these categories at the same time,” concluding that the aṅgas are “a mere description of literary types and not a division into water-tight compartments." (1959:11) 418 A 7.64/4:113,13 (SD 30.10); also in Vibhaṅga in terms of “analytic skill in Dharma”: (dhamma,paṭisambhidā) and “analytic skill in meaning” (attha,paṭisambhidā) (Vbh 294,22). 419 MĀ 1 (T1.421a17), T27 (T1.810a11) and EĀ 39.1 (T2.728c3). A Skt fragment parallel has preserved part of the aṅga listing; SHT III 878 R4, Waldschmidt 1971:127. 420 It has a Chin parallel in Ekottarika Āgama, EĀ 25.10 (T2.635a10), which compares mere learning of the 12 aṅgas to a cloud that thunders but does not rain, as stated in A 4.102. Another parallel, EĀ2 10 (T2.877b10), uses the same simile but does not mention the aṅgas. Here, one, like a cloud that thunders but does not rain, learns the suttas but does not understand the Dharma himself. http://dharmafarer.org 85 SD 58.1 The Oral Tradition of the Early Buddhists

attaining inner stillness; in other words, he is also a meditator. SD 44.4 Dhamma,vihārī Sutta 2 A 5.74/3:88 f (a shorter version of A 5.73) same teaching as preceding. SD 44.5

5.4.2.12 Such a contrast between mere learning or knowing, and true understanding is dramatically highlighted in the Alagaddûpama Sutta (M 22), a classic in the teaching of the spirit and the word of the Dharma. The Sutta opens with the recalcitrant monk Ariṭṭha obstinately holding on to a mistaken view that sexuality is not an obstacle to monastic training. So serious is his wrong view that his story is recorded in the legal case-histories of the Vinayas of different schools for the offence entailing pācittiya for such misconduct.421 According to the Sutta Commentary, Ariṭṭha is learned in the Dharma (MA 2:103,1). This suggests his conduct may be due to the arrogant belief that by mere learning he has fully understood the teachings. To highlight the seriousness of such folly and hubris, the Buddha relates the famous parable of the water- snake to remind us of mastering the 9 aṅgas, that is, both as true teaching and real purpose.422 According to the parable, a snake-catcher knows that if he were to grasp a water-snake by its tail it will fatally sting him. This illustrates the predicament of one who learns the teachings by way of the 9 aṅgas with neither examining their meaning nor understanding their true purpose; instead, one masters the teachings merely for the sake of winning debates with others. The wise snake-catcher grasps the snake by its neck, so that even when it coils around his hand, arm or limbs, he is still safe from it. So, too, it is with one who learn the 9 aṅgas properly knowing them and their true meaning and real purpose. In this context, the 9 aṅgas reflect the vicissitude of learning the teachings: it may have wholesome or unwholesome consequences, depending on our intentions. The same holds for the parallel versions to the Alagaddûpama Sutta in the Madhyama Āgama and the Ekottarika Āgama. The key difference is that these parallels, under the influence of later developments, is that they mention, not the 9 aṅgas, but a list of 12 aṅgas, popular in Mahāyāna.423

5.4.2.13 The Buddha Dharma is “endowed with meaning and phrasing” [5.4.2.11], and that is the way we should learn it; even more so, to teach it: that is, to do so fully in its spirit and its word. The Dhar- ma’s “spirit” entails understanding its language (nirutti) and mastering its meaning (attha); the Dharma’s “word” is the text of the teaching (dhamma) as well as our skill in teasing out its purpose with all our wits (paṭibhāna). These 4 qualities are so vital to learning and teaching the Dharma that they form a set of their own known as “the 4 analytic skills.” The 4 analytic skills (catu paṭisambhidā) are as follows:424

(1) the analytic skill in meanings (and purpose); attha,paṭisambhidā (2) the analytic skill in the teaching (and truth); dhamma,paṭisambhidā (3) the analytic skill in language;425 and nirutti,paṭisambhidā (3) the analytic skill in ready wit.426 paṭibhāna,paṭisam,bhidā Although these 4 analytic skills are the necessary qualities of a good Dharma teacher—and we may add, a sutta translator—clearly, from the suttas we have just examined [5.4.2.12], the first 2 qualities are

421 [62] The Dharmaguptaka Vinaya (T1428 @ T22.682a9), the Mahāsāṅghika Vinaya (T1425 @ T22.367a3), the Mahīśāsaka Vinaya (T1421 @ T22.56c12), the Mūlasarvāstivāda Vinaya (T1442 @ T22.840b21), with Skt and Tib parallels in Yamagiwa 2001:86,7, 87,9, the Sarvāstivāda Vinaya (T1435 @ T22.106a3), and Pali Vinaya (V 4:133,32). 422 M 22/1:133,24 (SD 3.13). 423 MĀ 200 (T1.764a14) and EĀ 50.8 (T2.813a16). 424 SD 28.4; SD 41.6 (2.2). 425 Further see Language and dfiscourse, SD 26.11. 426 See The Buddha’s humour, SD 98.1.

86 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists also those of a good Dharma student, indeed of anyone learning the Dharma so that they gain its fullest benefits.

5.4.3 From scripture to Scripture

5.4.3.1 As Buddhism grew in popularity, the monasteries, too, grew in size; more monastics were drowned in the world’s flow, attracting more of those who hoped to free-ride on these swirling currents of present benefits. In short, very few of them attain even streamwinning, not to say, arhathood. Procrustes- like, they stretched and cut, they dismembered the limbs of the teaching from the Dharma-body for their private purposes. They had mostly lost the spirituality of attainment and realization, and taken on a “modern” conception of authoritative texts or scripture (with the lower-case s), the sacred word in print. This new trend began soon after Aśoka’s time and the 1st century CE, with the rise of Christianity in Judea (in what is now Israel), and the rise of Mahāyāna in India. The Silk Road linked up Rome, Judea, Turkey, south and Central Asia to China, Korea and ended in Japan. The Kushans (the Yuezhi of the steppes), under Kanishka (c127-150 CE), sponsored Mahāyāna, alongside Greek cults, Hinduism and Zoroastrianism. He judiciously used religion to consolidate himself as Cakravarti, world ruler.427 The powerful needed religious harmony and prospered in it; the religions, too, prospered and evolved in this oikoumene, especially the Mahāyāna which grew great. Buddhism moved with the merchants on the Silk Road and spread all over Asia. It brought power and prosperity to the Sangha. In China, emperors followed Kanishka’s example and patronized Buddhism; in Korea Buddhism brought culture to the powerful; in Japan, the Buddhist priests had their own private armies and were feared even by the Emperor and the Shogun. Buddhism rose with empires and fell with them. When China, Korea and Japan saw the end of empires, Buddhism—as in India—also died out.

5.4.3.2 After Aśoka, the forest tradition lived on in the forest but in less numbers and significance. Buddhism was becoming more institutionalized, urbanized, powerful and wealthy (as it is becoming today). Its greatness climaxed in the rise of academic Buddhism in such universities as Nālandā (500 CE) and Vikramaśila a few centuries later. We are seeing the familiar rise of “modern Buddhism,” where tenet, titles and tribe matter more than early Buddhism. This period saw the rise of sectarian Buddhism, especially Mahāyāna and then Vajrayāna. The material wealth and worldly opulence of the Sangha drew the attention of the Muslim Turks during these golden centuries. Finally, the Buddhist world in India was annihilated, almost overnight, with the Turkish invasions. In 1192, Nālandā was sacked by Muslim troops of Mohammed Al-Ghauri; the library burnt for 3 months. In 1203, Vikramaśila was sacked by Bhaktiar Khalji, general of Qutb-ud-din Aibak, the monks massacred. Its last abbot, Śākyaśrībhadra, declared that Buddhism had been destroyed in India and fled to Tibet. By this time, Buddhism had spread beyond India northwards (to north and east Asia), and eastwards (to southeast Asia).

5.4.3.3 The Indian centuries before the Buddhist holocaust was a time of the rise of book Buddhism: the living oral tradition prospered into the Sacred Word; the Sangha became Church in the service of the State. Within Buddhism itself, the spiritual state described as āgama became Sūtras as statements of power. The singularity of the path of morality and samadhi became the plurality of Wisdom schools and ritual tribes. From this milieu arose the wealth of sectarianism, displacing the Buddha/arhat vision of awakening with the great vehicles of statues, status, lineages, images, empowerment and Dharma licence from an

427 On the cakka,vatti, see SD 36.9 (3); SD 36.10 (3). http://dharmafarer.org 87 SD 58.1 The Oral Tradition of the Early Buddhists external authority. In China, Buddhism was effectively sinicized and revised by the subtle yet profound influences of Daoism and Confucianism, which were themselves improved by this “barbarian” import. In India itself, the better-informed Buddhists were probably, inevitably, influenced by early Christianity. In AD 52, the Apostle Thomas landed in Muziris (Cranganore) on the Kerala coast, and established 7 churches (communities) there. Nestorian Christianity reached as far as China. Apparently, on account of such Christian influences, works like the Lotus Sutra were written and smuggled into the great new Bud- dhist gospel. In time, the lack of a unified idea of awakening, especially after the Buddha’s passing, each of the Buddhist sects or schools, unified only by language and culture, driven by dogma and power, began to have their own Scripture (with the big S): they rejected the historical Buddha or looked up to some other, external, glorified (mahā), God-like forms of enlightenment and theology. A key contribution to novel evolution was probably their prolonged mourning of the Buddha’s passing, nursed by the incredulous re- fusal to accept that the Buddha had died, that he could die. Perhaps, they were encouraged by early Christian notions of an eternal God and resurrection.

5.4.3.4 Mahāyāna, interestingly, arose around the same time as Christianity, and they were well linked with one another by the Silk Road [5.4.3.1]. Many Mahāyāna ideas parallel early Christian ideas— the Buddha is seen as docetic (nirmāṇa) like Christ as God taking human form; the Dharmakāya is close to the notion of Godhead; we only need faith or accepting Mahāyāna dogmas or practices (like reciting the Buddha’s name) to be assured of Paradise (such as Sukhāvati of the cosmic Buddha Amitābha). Indeed, it’s easier to understand such Mahāyāna theology through the Christian teachings than the basic teachings of early Buddhism: the one historical Buddha who dies (without resurrection); the Bud- dha’s awakening and that of the arhat are the same; and we can awaken only through self-effort [10.4.3]. Clearly, more study by scholars needs to be done in this fascinating area, and the academic study of Buddhism running out of our worthwhile or marketable theses to propose or defend, this comparative study of Buddhism and early Christianity is an almost certain trajectory to come.

5.4.4 The Chinese Buddhist canon

5.4.4.1 The legendary emperor Fu Hsi 伏羲 fúxī (2953-2838 BCE) was said to have invented writing using pictographs. The oldest surviving writing were ideographs etched on tortoise-shells, dating from 18th century BCE.428 Printing was invented in China between 680 and 750 (during the Tang dynasty), and its development was closely associated with Buddhism.429 This rich cultural background welcomed the Buddhist texts, even though the Chinese had a penchant for distrusting things foreign; but learning fascinated the pragmatic and adaptive Chinese. Hence, as early as the 1st century, the Chinese Buddhist canon started its evolution in China. The significance of the Chin- ese canon was that it became the basis for the east Asian Buddhist canon. In the 6th century, the Korean canon arose, and, starting from the 7th century, the Tibetan canon, which was finalized in the 14th century. The Japanese did not have their own Buddhist canon and used the Chinese one. It was only in the early 20th century that they produced the Taishō canon, and this became the international standard for the East Asian Buddhist canon to this day.430

5.4.4.2 Many of these Mahāyāna texts look or sound alike, but are often fundamentally different in mind and heart, like estranged siblings or lost cousins. However, insofar they look up to the Buddha’s

428 D Perkins, Encyclopedia of China, 1999:574 Writing system, Chinese. 429 D Perkins, Encyclopedia of China, 1999:398 f Printing. 430 See Wu & Chia, Spreading Buddha’s Word in East Asia, 2016:10 f.

88 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists teaching, when, in the face of extreme challenges or in profound moments of calm, they silently feel the need for living roots, seeing common teachings, related ancient history, the same blood flowing in different bodies. For example, the Chinese canon auspiciously opens its first 2 volumes comprising the āgamas (plural), which contain many of the sutras also found in the Pali canon, even if the Chinese texts are translations from the Sanskrit, not Pali, versions of these texts. Likewise, the Tibetan canon contains a great deal of Vinaya material that is thematically similar to those in the Vinaya of both the Chinese and the Pali canons, even if their texts are not exactly the same.

5.4.4.3 Historically, both the Chinese and the Tibetan canons include Mahāyāna sutras that are not found in the Pali canon. Further, the Tibetan canon includes many tantras that are not found in any other collection of Buddhist texts. Although from the Theravāda viewpoint, much of the Chinese and the Tibetan texts apocryphal or inauthentic (since they are not the Buddha’s word), much else significantly overlap, especially those in the 4 āgamas. In fact, this term āgama as used here refers to something very different from the āgama we have been discussing earlier on, that is, as found in āgatâgama [5.2.1.1]. This older sense, as found in the latter term, refers to a profound understanding of the Dharma based on personal realization.

5.4.5 The Āgamas in Chinese

5.4.5.1 Let us recap. We have mentioned that the early Buddhist oral tradition went through 3 basic stages of evolution—from āgama to aṅga to nikāya [5.1.1]—and we have also noted how āgama itself started as a reference to the state of awakening itself as authority: the early Buddhist texts issued forth from the mouths of the Buddha and the early arhats. Those noble ones who have attained awakening, or at least streamwinning, and understood the Dharma at their awakened level are said to be āgatâgama, “an attained one” [5.2.1.1]. The primary tasks of the true follower is to learn and master the Dharma from those who are attained (āgatâgama). Having mastered the teaching in theory (pariyatti), they must now go on to put it into practice (paṭipatti), that is, train in mental solitude, purifying, focusing and freeing the mind, so that they themselves attain realization (paṭivedha).431 However, with the growing number of unawakened monastics and more lay followers, the number of those who learned and mastered Dharma in theory also grow, but they do not awaken commensurately. They have mastered the āgama but they have not attained any state of the path, not even streamwinning. On their account, āgama takes on a new sense knowing the Dharma without any attainment.

5.4.5.2 In a way, many of us who know the Dharma well today have reached this stage, too, that is mastered the āgama as text. This is certainly better than not knowing the Dharma at all; but it is even better when we keep up our training in moral virtue, as the basis for the training in concentration, or at least mindfulness, so that we attain some level of insight wisdom. It is in this spirit, that the teaching of aṅga, the limbs of the teaching is used, as highlighted dramatically in the Alagaddûpama Sutta (M 22): the good that we have learned we should put into practice for reaching the path [5.4.2.12]. The term aṅga refers any part or aspect of the Dharma which are suitable or appropriate for teaching others, and for the student to learn for themselves. While the āgama refers to

431 The triad of study (pariyatti), practice (paṭipatti and realization (paṭivedha), is a Comy set known as the “3 good states” (saddhamma) (VA 2235; AA 5:33): see The levels of training, SD 40a.4 esp (2); Notion of diṭṭhi, SD 40a.1 (3.4). http://dharmafarer.org 89 SD 58.1 The Oral Tradition of the Early Buddhists the whole of the Buddha Dharma, aṅga refers to a more manageable, more gradual¸ approach to learning and training for the path.

5.4.5.3 It is difficult to know exactly when the monks started to assemble the various suttas and texts of the Buddha’s teachings into some kind of structured collection. Thanks to the Chinese penchant for being meticulous about dates—we know the exact year for the translations of each of the Chinese Āgamas [5.4.5.4]—the whole of the Chinese Āgamas were translated between the 4th and 5th century CE. Over 500 years before that, the Pali Tipiṭaka—the Vinaya, the 5 Nikayas (suttas) and the Abhidhamma—had already reached Sri Lanka during Aśoka’s time (268-232 BCE).

5.4.5.4 Also included as “early Buddhist texts” are the translations or parallels of these Nikāya texts in Chinese, Tibetan, Sanskrit and other Indian dialects. The first 2 volumes of the 100-volume Chinese Tripiṭaka—大藏經 dà zàng jīng432—preserves valuable early texts that correspond to the 4 Pāli Nikāyas, though they are not as complete as the latter. As a whole, however, the Chinese Āgamas are roughly coextensive with the Pali Nikyas. The 4 Āgamas are, respectively, as follows:433

(1) Dīrgha Āgama434 長阿含經 cháng āhán jīng “the long discourses” T1 (2) Madhyama Āgama 中阿含經 zhōng āhán jīng “the medium discourses” T26 (3) Saṁyukta Āgama 雜阿含經 zá āhán jīng “the connected discourses” T99 (4) Saṁyukta Āgama 別譯雜阿含經 biéyì zá āhán jīng “the alternate connected discourses” T100 (5) Ekôttar(ik)a Āgama 增壹阿含經 zēngyī āhán jīng “the numerical discourses” T125

(1) Dīrgha Āgama (DĀ), 長阿含經 cháng āhánjīng, “the long discourses,” This is T1 (22 rolls, 30 sutras), Dharmaguptaka affiliation, translated by Zhu Fonian 竺佛念 zhú fóniàn, based probably on Prakrit read by Buddhayaśas 佛陀耶舍 fótuóyēshě, 15th year of Hung Shih 弘始 hóng shǐ, 413 CE, Late Qin 後秦 hòuqín (384-417). Other translations are found in T2-25.435 Large parts of the (Mūla)Sarvāstivādin Dīrgha Āgama survived in substantial Sanskrit fragments.436 Versions of some long discourses not belonging to the (Mūla)Sarvāsti- vāda are extant in Indic-language fragments. Portions of the text survive in Tibetan translation.

DĀ corresponds to the Dīgha Nikāya of the Pali canon [6.2.4.1]. The Chinese DĀ translation (above) contains 30 sūtras in contrast to the 35 suttas of the Pali Dīgha Nikāya.437 Modern textual research has re-

432 The early sutras (in translation), called Āgama Sūtras or simply Āgama (Chin 阿含經 āhánjīng; Jap 阿含部 Agon- bu); see Enomoto 1986 (below). They are preserved in vols 1-2 (order 1-151) of the 100-vol Taishō Shinshū Daizōkyō (大正新脩大藏經 dàzhèng xīnxiū dàzàngjīng) (The Taishō Revised Tripiṭaka): https://21dzk.l.u-tokyo.ac.jp/SAT/index_- en.html. On the Chin Buddhist canon, see SD 40b.2 (2.6.4). On various versions of the Chin Tripiṭaka, see SD 52.2 (1.2.3.1 n). On their origins: Enomoto, “On the formation of the original texts of the Chinese Āgamas,” BSR 3,1 1986: 19-30. 433 See A Hirakawa (tr P Groner), “Buddhist literature: Survey of texts,” (ed Eliade & Adams), The Encyclopedia of Religion 1987:513; J-U Hartmann, “Āgama/Nikāya,” (ed Buswell,) Ency of Buddhism, 2004:10-12; Analayo, “Āgama- /Nikāya,” Brill’s Ency of Buddhism, 2015a:50-59. 434 These names may also be spelt as a compound, respectively, thus: Dīrghâgama, Madhyamâgama, Saṁyuktâga- ma, and Ekottarâgama or Ekottarikâgama. 435 For full list of DĀ (T1), see K648 (KBC 1979:349-358). 436 Hartmann, “Contents and structure of the Dīrghāgama of the (Mūla)Sarvāstivādins,” 2004b; 20th-cent discovery of fragments, Allon, ““Recent discoveries of Buddhist manuscripts from Afghanistan and Pakistan and their significance,” 2008.

90 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists vealed the Dīrgh’āgama to comprise a total of 47 sutras in Sanskrit, of which 36 have known Pali versions, with 10 other Dīgha suttas whose parallels are found instead in the Madhyam’āgama.438 The 30 DĀ sutras have these 27 Pali parallels, listed here as “DĀ (D)”: DĀ 1 (D 14 Mahā’padāna), 2 (16 Mahā,parinibbāna), 3 (19 Mahā Govinda), 4 (18 Jana,vasabha), 5 (27 Aggañña, also MĀ 154), 6 (26 Cakka,vatti Sīha,nāda, also MĀ 70), 7 (23 Pāyāsi, also MĀ 71), 8 (25 Udumbarikā Sīha.nāda, also MĀ 104), 9 (33 Saṅgīti), 10 (34 Das’uttara), 11 (nil), 12 (nil), 13 (15 Mahā,nidāna, also MĀ 97), 14 (21 Sakka Pañha, also MĀ 134), 15 (24 Pāṭika), 16 (31 Sigāl’ovāda), 17 (29 Pasādika), 18 (28 Sampasādanīya), 19 (20 Mahā Samaya; S 1,4,7 Samaya), 20 (3 Ambaṭṭha), 21 (1 Brahma,jāla), 22 (4 Soṇa,daṇḍa, 23 (5 Kūṭa,danta), 24 (11 Kevaṭṭa), 25 (8 Kas- sapa Sīha,nāda), 26 (13 Te,vijja), 27 (2 Sāmañña,phala), 28 (9 Poṭṭha,pāda), 29 (12 Lohicca), 30 (nil). DĀ makes these 7 omissions of Pali suttas: D 5 Mahāli, 7 Jāliya, 10 Subha S, 17 Mahā Sudassana (MĀ 68), 22 Mahā Sati’paṭṭhāna, 30 Lakkhaṇa (MĀ 59), 32 Āṭānāṭiya. These 7 omitted suttas were probably not in the common early canon, but were added to D after the schism (during the sectarian period). The following 3 DĀ sutras (in Chinese) (marked as “nil” in the list above) have no Pali parallels: DĀ 11 增一經 zēng yī jīng, DĀ 12 三聚經 sān jù jīng, DĀ 30 世記經 shì jì jīng. They were probably added later.

(2) Madhyama Āgama (MĀ), 中阿含經 zhōng āhánjīng, “the medium discourses,” According to the oldest extant Chinese catalogues,439 there are 2 MĀ’s: 1. by Saṅghadeva, and 2. by Dharmanandī (from Tokhara) Zhú Fóniàn. The latter is the earlier one, and will be mentioned first. (a) The translation by Zhú Fóniàn 竺佛念 and Dharmanandī or Dharmanandin 曇摩難提 tánwúnántí (from Tokhara) was done in 384-385. Apparently, this was replaced by the following. (b) T26 (60 rolls, 222 sutras), associated with the Sarvâstivāda school, translated by the Kashmiri monk, Gautama Saṅghadeva 瞿曇僧伽提婆 qútán sēngqié típó, between the 10th day, 11th moon, 1st year and the 25th day, 6th moon, 2nd year, of Lung An 隆安 lóng ān, 15 Dec 397-24 Jul 398, Eastern Jin dynasty 東晉朝 dōng jìn cháo (266-420) in Tung-t'ing Monastery 東亭寺 dōng tíng sì. The traditional Chinese records report that Saṅghadeva’s translation was based on Prakrit440 read out to him by the Kashmiri Saṅgharakṣa, the scribe was Dàocí 道慈, who was assisted by Libao 李寶 lǐ bǎo and Kànghuà 康化 kāng huà (T1.809b26). Understandably, this was a better translation of MĀ, but we do not know who edited or revised it.441 (c) Other individual translations are preserved in T27-94; and a few exist in Tibetan transaltion.

MĀ corresponds to the Majjhima Nikāya of the Pali canon [6.2.4.2]. However, the Chinese MĀ translation has 222 sūtras, in contrast to the 152 suttas of the Pāli Majjhima Nikāya. This larger MĀ number is due to its translations including a large number of suttas from the Aṅguttara (79 suttas), and also Saṁ- yutta (12), Dīgha (10), Vinaya (4), Iti,vuttaka (3), Thera,gātha (3), Sutta Nipāta (2), Jātaka (2), Divyāvadāna (1). A number of the individual MĀ sutras contain the combined translations of more than 1 sutta or title. Sanskrit fragments of the Sarvāstivāda Madhyama Āgama have been found (such as at Turfan in Cen- tral Asia). Sections of it also survive in Tibetan translation. The MĀ preface contains the term tathāgata,- garbha (“the thus-come’s womb,” popularly “Buddha-nature”). Suggesting a nascent influence of the Mahāyāna.

437 Hartmann, “Zu einer neuen Handschrift des Dīrghāgama,” 2000, “Another addition to the An Shigao corpus?” 2002; Salomon, “Recent discoveries in early Buddhist manuscripts,” 2006:356. 438 R S Bucknell, “The structure of the Sanskrit Dīrgha-āgama from Gilgit vis-à-vis the Pali Dīgha-nikāya,” 2014:70- 101, esp Table 2 for a collation of D/DĀ (Skt) /MĀ (Chin) parallels. Also Analayo 2015a:50-52. 439 出三藏記集 chū sānzàng jì jí (T2145.10b-c); 衆經目錄 zhòng jīng mùlù (T2146.129a. 440 See Analayo, Madhyama-āgama Studies, 2012:516-521. 441 See Analayo, ib, 2012:1 n1. http://dharmafarer.org 91 SD 58.1 The Oral Tradition of the Early Buddhists

Of the 222 MĀ sutras, 99 have Majjhima (M) parallels. Other Pali parallels are as follows:442 Dīgha 14 suttas; Saṁyutta 20 suttas; Aṅguttara 89 suttas; Thera,gāthā 5 parallels; Sn 2 suttas; U; It 3 suttas; V 5 parallels. 19 MĀ sutras have no Pali parallels. The MĀ sutras have the following 106 Pali parallels, with 100 Majjhima parallels and 6 Dīgha parallels. The MĀ number is listed first followed by the Pali sutta number within (round brackets).443 MĀ 9 (M 24), 10 (2), 14 (61), 19 (101), 26 (69), 27 (97+D 15), 28 (143), 30 (28), 31 (141), 32 (123), 34 (124), 63 (81), 64 (130), 67 (83), 71 (D 23) 72 (128), 75 (105), 77 (68), 78 (49), 79 (127), 81 (119), 85 (113), 86 (148), 87 (5), 88 (3), 89 (15), 91 (8), 93 (7), 97 (D 15), 98 (10), 99 (13), 100 (14), 101 (20), 102 (19), 103 (11), 104 (D 25), 105 (6), 106 (1), 107-108 (17), 115 (18), 131 (50), 132 (82), 133 (56), 134 (D 21), 135 (31), 144 (107), 145 (107), 146 (27), 150 (96), 151 (93), 152 (99), 153 (97), 154 (D 27), 161 (91), 162 (140), 163 (137), 164 (138), 165 (133), 166 (134), 167 (132), 168 (120), 170 (139, 135), 171 (136), 173 (126), 174 (45), 175 (46), 178 (25), 180 (142), 181 (115), 182 (39), 183 (40), 184 (32), 185 (31), 186 (47), 187 (112), 189 (117), 190 (121), 191 (122), 192 (66), 193 (21), 194 (65), 195 (70), 196 (104), 198 (125), 199 (129), 200 (22), 201 (38), 203 (54), 204 (26), 205 (64), 206 (16), 207 (77), 208 (79), 209 (80), 210 (44), 211 (43), 212 (90), 213 (89), 214 (88), 216 (87), 217 (52), 221 (63). The following MĀ sutras have multiple parallels in the Nikāyas: MĀ 18 A+V, 21 Ax2, 36 (D+A), 37 (A+V), 45 Ax2, 52 Ax2, 56 A+U, 58 D+S, 64 D+A, 83 A+Tha, 98 D+M, 105 M+A, 113 Ax2, 116 A+V, 122 Ax2, 123 A+Tha, 137 A+It, 138 S+A, 140 S+It, 141 Sx10 + It, 142 D+A, 160 Ax2, 169 Ax2, 170 Mx2, 188 Ax2, 217 M+A. 17 omissions: MĀ 7, 39, 44, 46, 54, 55, 65, 69, 80, 92, 147, 159, 176, 177, 218, 219, 222.

(3) Saṁyukta Āgama (SĀ), 雜阿含經 zá āhánjīng, “the connected discourses,” (a) T99 (T2.1a) (50 rolls, 1362 sutras),444 belonging to the (Mūla)Sarvâstivāda school,445 with a partial translation belonging to the Kāśyapīya school.446 Working in Chang’an 長安 cháng'ān, the translation was done by Baoyun 寶雲 bǎo yún, from the San- skrit read out by Guṇabhadra 求那跋陀羅 qiúnà bátuóluó (394-468), between 12th and 20th years of Yüan Chia 元嘉 yuán jiā (435-436 or 443)447 at Waguan Monastery 瓦官寺 wǎ guān sì, Yangdu (楊都) yáng dū, in the Song state 劉宋 liú sòng (435-443).448 (b) There is also an incomplete SĀ Chinese translation (SĀ2) 別譯雜阿含經 bié yì zá āhānjīng, “the alternate connected discourses,” T100449 (20 rolls,450 364 sutras)451 of probably the Kāśyapīya (飲光部 yìn guāng bù) school,452 by unknown translator/s of northern China, from around the Three Qin (三秦 sān qín)

442 Key: A = Anguttara Nikāya; D = Dīgha Nikāya; It = Iti,vuttaka; M = Majjhima Nikāya; S = Saṁyutta Nikāya; Tha = Thera,gāthā; U = Udāna; V = Vinaya. 443 Key: MĀ = Madhyama Āgama sutra; M = Majjhima Nikāya sutta; D = Dīgha Nikāya sutta. 444 T99.2.373b07. KBC K650 (1979:376-450) gives a total of 1507 titles. SHI Yinshun 1983 has restored the SĀ sutra sequence, based on discoveries by LÜ Cheng (1923). See Bingenheimer, “The shorter Chinese Saṁyukta Āgana,” 2006:21+n2. 445 It seems possible that the MS of this collection was acquired by the pilgrim Fa-hsien 法顯 fǎxiǎn (c 337-422) in Sri Lanka, where he stayed in the Abhayagiri monastery (Glass, “Guṇabhadra, Bǎoyún, and the Saṁyuktāgama,” 2010:200). 446 This is from the Taishō printed ed. However, 3 of these are interpolations of other texts and do not belong to SĀ proper: A Glass, Four Gāndhārī Saṁyuktāgama Sutras, 2007:28 + n11. For discussion of school affiliation, See Glass op cit 2007:28 + n10. 447 The dating 435-436 is based on Enomoto (2002b:32). Other scholars date the tr more broadly to 435-443. For a detailed discussion T99-101, see A Glass 2007:38-50 (1.3). 448 For full list of MĀ (T99), see K650 (KBC 1979:376-450). 449 For full list of SĀ2 (T100), see K651 (KBC 1979:450-458). 450 The Taishō gives 16 rolls, but this prob an error arising from the earlier confused state of the text. Hence, it should not be thought that 4 rolls are missing. Nanjio’s catalogue give 20 rolls: Mizuno 1988:11-2. On the reconstruc- tion of the T99, 18 rolls correspond with T100 (Mukai 1985:18): see Glass 2007: 1.3.2.

92 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists period, 352/385-431 CE.453 A comparison of the Sarvāstivādin, Kāśyapīya, and Theravadin texts reveals a considerable consistency of content, although each recension contains texts not found in the others.454 A 3rd SĀ translation is T101, the earliest of the 3, is actually a partial translation, an anthology of 27 short sutras, in 1 roll, probably translated by An Shigao 安世高 ān shìgāo (fl c148-168), and may also belong to the Sarvāstivāda. Its sequence and content do not closely parallel T99. Comparing with other versions of the sutras suggests that they came from 5 sections of a more complex text. Two of the sutras are the same as An Shigao’s translation of T2.150a, and these, along with a third, may have come from his translation of EĀ and mistakenly put into T101.455 There are also a number of independent translations of single suttas in the Taishō canon, and these are numbered T102-124.456 Portions of the Sarvāstivāda Saṃyukta Āgama also survive in Sanskrit457 and Tibetan translation, some of which form a collection entitled Mahāsūtra (The Great Sutras).458

SĀ corresponds to the Saṁyutta Nikāya of the Pali canon [6.2.4.3], which contains over a thousand short suttas organized by topic. The texts, however, were somewhat corrupted during their transmission in China, for example, an extraneous passage from a later life of Aśoka was mistakenly included. In the 20th century, a series of scholars reconstructed SĀ’s original sequence.459 In 2014, The Collation and Annotation of Saṁyuktāgama《雜阿含經校釋》zá āhánjīng jiàoshì (Chin- ese) was published by Wang Jianwei and Jin Hui was published in China.

(4) Ekôttara Āgama (EĀ), 增壹阿含經 zēngyī āhánjīng “the numerical discourses” This is T125 (473 sutras),460 of uncertain provenance: probably Dharmaguptaka or Sarvāstivāda, or, less likely, to Mahāsaṅghika school or its offshoot, the Prajñaptivāda school. It was translated by Zhu Fo- nian 竺佛念 [above] probably based on a Prakrit recited by Dharmanandī 曇摩難提 [above] in Fu Qin state 苻秦 fúqín or Eastern Chin dynasty 東晉 dōng jìn in 384; then edited by Gautama Saṅghadeva, 397-398: 1st year of Lung An 隆安 lóng ān in Lu-shan 廬山 lú shān. Various individual translations are preserved as T126-149.461 There is also a partial translation (T150a, an early one) done by An Shigao 安世高 [above].462 Parts of a collection of numerical discourses are extant in Sanskrit (Tripathi, 1995), and also other fragmentary parallels and a few counterparts in Tibetan trans-

451 T100.2.491c26. This number represents only 2 major divisions of the full collection; prob an incomplete tr of a longer work rather than an anthology (Anesaki 1908, 3, 75). 452 Other scholars think that it belongs to the Mahīśāsaka, Dharmaguptaka, Sarvāstivāda or Mula,sarvāstivāda: Glass 2007:28 n14. 453 Oxford DB 2003:250 Saṁyukta Āgama. The dates: 352-431 are from Anesaki, “The 4 Buddhist āgamas in Chin- ese,” 1908:3, and 385-431 Mizuno, ‘Betsu agon kyō ni tsuite,’ 1970:486. Mayeda thinks that it may be slightly earlier than T99 (1985:100). 454 See A Glass, Four Gāndhārī Saṁyuktāgama Sutras, 2007:28 f. 455 See http://buddhistinformatics.chibs.edu.tw/BZA/. See Harrison 1997:265, 276, 2002:11, 19. 456 See Glass 2007:28 f. 457 Tripathi, Fünfundzwanzig Sūtras Des Nidānasaṃyukta, 1962. 458 P Skilling, Mahāsūtras, 1994. 459 On the misplacements of 2 SĀ fascicules–nos 23 (SĀ 604) + 25 (SĀ 640, 641)–and their subsequent reconstruct- ion, see Analayo 2015a:56. 460 Unlike the other 3 Āgamas, EĀ is “highly erratic, and internally inconsistent, possibly being an unfinished draft” (Sujato & Brahmali 2015:41). It incl Mahāyāna additions, giving it a late date of completion (Lamotte 1988:154, 156). 461 For a full list of EĀ (T125), see K649 (KBC 1979:358-376). 462 Harrison, “The Ekottarikāgama Translations of An Shigao,” 1997. http://dharmafarer.org 93 SD 58.1 The Oral Tradition of the Early Buddhists lation. A commentary on the first part of the Ekottarikāgama is found in Fēnbié gōngdé lùn 分別功德論 “Treatise on the Kinds of Merit” (T1507). EĀ corresponds to the Aṅguttara Nikāya of the Pali canon [6.2.4.4]. However, of all the Āgamas, EĀ has considerable disparity between the Pali and the Sarvāstivādin version. More than 2/3 of the sutras found in one but not in the other, “which suggests that much of this portion of the Sūtra Piṭaka was not formed until a fairly late date.”463

5.4.5.5 According to A K Warder, EĀ references 250 Prātimokṣa rules for monks, which agrees only with the Dharmaguptaka Vinaya, which is also located in the Chinese canon. He also views some of the doctrine as contradicting tenets of the Mahāsaṅghika school, and states that they agree with Dharmagup- taka views currently known. He therefore concludes that the extant Ekottara Āgama is that of the Dharma- guptaka school.464 Of the 4 Āgamas in the Chinese Buddhist Canon, the Ekottara Āgana is the one which differs most from the Pali suttas. It contains variants on even such standard teachings as the noble eightfold path.465 According to Oxford Dictionary of Buddhism, “there is considerable disparity between the Pāli and the [Chinese] versions, with more than two-thirds of the sūtras found in one but not the other compilation, which suggests that much of this portion of the Sūtra Piṭaka was not formed until a fairly late date.”466

5.4.5.6 Clearly, the Buddhist missionaries and translators in China used the term āgama in the later sense of “text, scripture.” Around the same time, Buddhaghosa, too, in his Pali Commentaries, used the word āgama interchangeably with nikāya [6.2.1.2]. Clearly, the usage of the term āgama in this sense was a later development with the rise of Sanskrit cultural hegemony. At least, the Chinese Buddhist used the term āgama for their received texts that preserved the early teachings, that is, the 4 āgamas. The equivalence or parallel between these collections and the Pali canon is often remarkably close, attesting to a common ancient source, an urtext. However, such correspond- ence are not always exact, but often enough, too, there are also divergences. Further, it is not known how many of these schools had an equivalent to the the Khuddaka Nikāya, the 5th Nikaya of the Pali Canon, since none survives in Sanskrit, and none were translated into Chinese. The Chinese canon also preserves, in translation, individual texts of various Vinaya texts, and other San- skrit and Prakrit texts, including parts of the Arthapada Sūtra 義足經 yì zú jīng (T198, Aṭṭhaka,vagga, “the octet chapter” of the Sutta Nipāta),467 the Dhammapada (T210,468 211, 213), and the Udāna,varga (the Sanskrit version of the Dhammapada) (T212).469 Although these āgama texts were often originally in Sanskrit, it is unlikely that there was ever a single Sanskrit canon, since the Buddha forbids its use in learning or teaching Dharma, though later traditions, especially the Mahāyāna, chose to ignore this ruling.470 Most of the early 18 schools471 like the Sarvastivad-

463 Oxford DB 2003: Ekottara Āgama. 464 Warder, Indian Buddhism, 2000:6. 465 Sujato, “About EĀ”: https://sites.google.com/site/ekottara/about. 466 Oxford DB 2003: Ekottara Āgama. 467 Tr Zhi Qian 支謙 zhī qiān from a Prakrit text, after 229 CE. Eng Tr P V Bapat, “The Arthapada-sūtra,” 2 parts, 1945+50. https://arthapada.blogspot.com/p/introduction.html. 468 法句經 fǎ jù jīng (T210) was tr c220 by the Indian monk Wei-ch’i-nan 維祇難 wéi qí nán and others; it preserves the early format of the text. 469 T210, 211, 212, 213 and T198 are found in T4 (Taisho vol 4). Further for other texts, incl those on the Buddha’s life in Chinese, see Macmillan Ency Rel (ed Eliade & Adams) 1987:513. 470 See SD 26.11 (3.1.4). 471 The 18 early Buddhist schools (nikāya): (1) Sarvāstivāda: 1. Mūla,sarvāstivāda; 2. Kāśyapīya; 3. Mahīśāsaka; 4. Dharma,guptaka; 5. Bahu,śrutīya; 6. Tāmra,śāṭīya; 7. Vibhajya,vādin (P vibhajja,vāda); (2) Saṁmitīya; 8. Kauru,kulla-

94 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists ins and the Lokottaravādins (a branch of the Mahāsaṅghikas), had their own canon in Sanskrit. Others might have had canons in various forms of Prakrit or Middle Indian, such as Gāndhārī, some of which have been uncovered in recent times [6.2.4.1].

5.4.5.7 Although the Chinese Āgama texts came from various schools—even because they came from different sources—they provide a valuable source for comparison with the Pali Nikāya, enabling us to discern pre-sectarian textual strata, and, even today, are valuable for source-critical and comparative studies of the Pali suttas themselves. Although much of these texts have undergone the vicissitudes of transmission and translation between two culturally disparate languages, and a religion alien to the Chinese, they are often helpful in giving us broader perspectives and clearer insights into the early teachings. In the early 20th century, numerous fragments of Sanskrit sutra manuscripts were found in Central Asia, enabling scholars to recover at least a small part of the Sūtra Piṭaka of the (Mūla)Sarvāstivādins. Later, among the Gilgit finds were fragments of the Ekottarika Āgama of the same school. Manuscript finds from Afghanistan and Pakistan also contained many sutra fragments from the scriptures of at least two schools, the (Mūla)Sarvāstivādins and probably the Mahāsaṅghikas. Most notable among them is a manuscript of the Dīrgha Āgama of the (Mūla)Sarvāstivādins. Unlike colophons of Vinaya texts, those of single sutras or sutra collections never mention schools, which often made it difficult to ascribe the provenance. Hartmann concludes that school affiliation of Āgama texts may have been less important than modern scholars tend to believe,472 which means that these early schools probably had some kind of common canon, at least shared by some of them.

6 NIKĀYA: THE SUTTA COLLECTIONS

6.1 ĀGAMA AND NIKĀYA

6.1.1 North and south

6.1.1.1 As we have noted [5.4.5], by the time Buddhism reached China, and the Buddhist texts began to be translated into China, the terms āgama 阿含經 āhán jīng was (and still is) used for them. In other words, the term āgama had, by then, evolved in meaning from “attainment (of the path)” [5.2.1.2] to mean simply “text” as authoritative scripture or simply Scripture. Āgama retained its basic meaning of “(received) tradition,” in the sense of preserving the early canonical text. It was also a time that saw the beginnings of advent of the Mahāyāna teachings in India (in the early Common Era), and it was spreading in China alongside the early teachings. For various reasons, the Mahāyāna teachings appealed to the Chinese more than the early Buddhist teachings. In fact, we see Mahāyāna teachings and texts evolving amongst the Chinese Buddhists, too.473

6.1.1.2 However, some 5 centuries before Buddhism first reached China, these same ancient texts— the 4 āgamas that were translated into Chinese—had reached Sri Lanka, and presumably also found in much of South India, even as far south as the Tamil regions, such as Kanchipuram (Kañcīpuram) district; it is also the name of a temple town, Tamil Nadu’s capital, located just southeast of Madras or modern Chennai, from where the famed commentator, Dhamma,pāla (who lived just after Buddhaghosa’s time) came [4.1.1.3]. ka; 9. Avantaka; 10. Vātsī,putrīya; (3) Mahā,sāṁghika: 11. Pūrva,śaila; 12. Apara,śaila; 13. Haimavata; 14. Lokottara,- vāda; 15. Prajñapti,vāda; (4) *Sthavira,nikāya: 16. Mahāvihāra,vāsa; 17. Jetavanīya; 18. Abhayagiri,vāsa. 472 See Hartmann, “Āgama/Nikāya,” in Ency Bsm (ed Buswell) 2004a:11. 473 For a detailed study of the Sinicization of Buddhism, see How Buddhism became Chinese, SD 40b. http://dharmafarer.org 95 SD 58.1 The Oral Tradition of the Early Buddhists

Another famous Buddhist scholar, was Kassapa (fl c1200 CE), who wrote a commentary on the Abhi- dhamma mātikā in his Mohā,vicchedanī, came from Chola (or Coḷa Nadu), which encompasses the lower reaches of the Kaveri river and its delta. It was the country of the powerful thalassocratic Chola dynasty, the world’s longest reigning dynasty (3rd century BCE-1279), whose power, at its height, controlled much of south India and almost all of Sri Lanka from the 9th to 13th centuries, and even invaded Srivijaya (including Kedah, peninsular Malaysia) in the early 11th century. All this shows the significant presence of Buddhism in south India (including Andhra Pradesh, with its great stupa at Amaravati, bordering Tamil Nadu on the north). However, Buddhism disappeared from south India, too, following the Buddhist holocaust in the north. It could not have been due to the Turk invasion, which did not reach that far south. Further research and study are needed here to learn about the disappearance from south India.

6.1.1.3 These ancient texts, roughly the same historical age as the Āgamas of the Chinese, were known as nikāya. Like, āgama, the term nikāya, too, denote a division of the ancient Buddhist canon of suttas, that is, the fourfold collection of long collection (dīgha,nikāya and dīrgha āgama), middle-length collection (majjhima,nikāya and madhyama āgama), connected collection (saṁyutta nikāya and saṁyuk- ta āgama), and (aṅguttara nikāya and ekottara āgama). While the Āgamas were preserved in the first 2 volumes of the Chinese canon, the Nikāyas grew into the fivefold collection known as the sutta collection (sutta piṭaka) in “3 baskets” (ti,piṭaka) of Vinaya, Sutta and Abhidhamma Piṭakas.

6.1.2 Spread of the early canon

6.1.2.1 After Buddhism has arrived and established itself in a country or culture, the religious elite with support from the social gentry would patronize and promote it. In no time, they started to put together their own Buddhist canon, a process that often takes centuries. Historically, the Buddhist canon was put together by the Buddhist council held in Pāṭali,putra, under the auspices of emperor Aśoka—traditionally known as the 3rd council—convened in the 17th year of Aśoka’s reign, around 240 BCE. This was also the time when the Pali canon was closed; that is, no more additions or revisions were made to it. Although this was probably only the affair of some community of traditional Buddhist elders, it was of historical significance for the Theravāda. 6.1.2.2 It was this significant presence of traditional Buddhism in Aśoka’s realm that led to the advent of Buddhism in Sri Lanka. It was basically an exchange between kings, those of India (Aśoka, r 268-232 BCE) and of Anuradhapura, in north-central Sri Lanka (Tissa, r 247-207 BCE). King Tissa later adopted the title devānam,piya, “beloved of the gods,” Aśoka’s title, to celebrate their new alliance that brought Bud- dhism to Anuradhapura as well as consolidated his kingship. Hence, by 250 BCE, the Buddhist canon (including the Commentaries) had arrived in Sri Lanka. This key event was a vital start for the spread of what we today know as the Pali canon or Tipiṭaka [6.2]. In due course, the canon arrived in Burma (modern Myanmar), Siam (modern Thailand), Khmer (modern Cam- bodia) and Laos. Of course, it is likely that Myanmar received the Pali canon from India, too, on account of her proximity of India.474

474 On “Nikāya Buddhis” (incl the early schools), see Routledge Ency of Buddhism 2007:549-558.

96 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

6.2 Ti,piṭaka

6.2.1 Oral transmission

6.2.1.1 It is not known when redactors started to put together individual suttas into structured collections. According to tradition, the suttas were already collected together by the time of the 1st council, held 3 months after the Buddha’s death in order to establish the suttas as the authentic Buddha-word (buddha,vacana). Scholars, however, tend to see the texts as continuously growing in number and size from an unknown core or urtext, and, in the process, undergoing changes in language and content. During the early after- centuries, the texts (the suttas and the Vinaya) were orally transmitted, for the proper preservation of the texts. We have mentioned a method of categorizing the texts into 2, 3, 4, 9 or 12 “limbs” (aṅga), but this simply refers to the kinds of teaching, rather than a scheme for classifying the teachings [5.4]. The aṅga scheme was replaced by a new idea of āgama, which exudes a sense of authenticity, since it originally refers to the teachings of those who have attained the path. [5.2.1]

6.2.1.2 The early Buddhists and those who upheld and passed on the early teachings did not use the term āgama in the manner the Chinese used for their early texts preserved in their canon. However, the Pali commentarial tradition saw āgama and nikāya as overlapping significantly: both terms refer to “scriptural texts.” Even in the key term āgatâgama (āgata + āgama) was redefined to mean “that which has come down from the āgamas,” which retained the sense of the attained (those who have reached the path, especially the arhats). Yet, it also has taken on a new sense: it is also the texts that has been handed down from them. The term used for this is nikāya. Hence, the Commentaries speak of āgatâgama as being “those who have mastered these Āgamas, that is, the Dīgha (Nikāya) and so on” (dīgh’ādiso yo koci āgamo āgato etesan’ti āgatâgamā (AA 3:382,7); The 5 Āgamas are called the 5 nikāyas; one becomes an expert in coming into mastery of these (nikāyas) —one is called āgatâgama (pañca,nikāyā pañca āgamā nāma; tesu āgamesu yesaṁ eko pi āgamo āgato paguṇo pavattati, te āgatâgamā nāma (AA 2:189.17).

6.2.2 The reciters

6.2.2.1 In due course, certainly by Aśoka’s time [6.1.2.2], the suttas were apportioned into large “collections” (nikāya), more or less arranged by length, that is, long texts (dīgha,nikāya), middle-length texts (majjhima,nikāya), connected texts (saṁyutta,nikāya) and numerical texts (aṅguttara,nikāya). Those monastics who specialized in memorizing and reciting these texts—called bhāṇaka—were known by the col- ection or text that they memorized, that is, respectively, the dīgha bhāṇaka (the long text reciters), the majjhima bhāṇaka (the middle-length text reciters), the samyutta bhāṇaka (the connected text reciters) and the aṅguttara bhāṇaka (the numerical text reciters).475

6.2.2.2 None of these texts (at least as we have them now) are the ipsissima verba or “original words” of the Buddha alone. At best, they are records, like transcriptions, of what the Buddha or the early arhats have taught. Like all transcriptions, they are edited transcriptions. If we only had the words of the Buddha, we would surely have great difficulty understanding it: it would be like listening to a recorded talk that has no introduction, no title, no context.

475 See Norman 1983b:8 f. http://dharmafarer.org 97 SD 58.1 The Oral Tradition of the Early Buddhists

Fortunately, the ancient reciters (bhāṇaka), monastics who lived in groups and specialized in memorizing individual Nikāyas,476 put together the teachings of the Buddha and early saints in a recitable manner, very often giving their contexts (the venue, the audience, the occasion, etc). Most usefully for us, the core of such texts and collections, goes back to early Buddhism, even to the Buddha himself.477

6.2.2.3 Technically, the term anussavana refers to the “aural tradition,” especially that of the brahminical tradition, while mukha,pāṭha478 refers to the “oral recitation” of the Buddhist reciter [1.2.3.1]. The difference is significant. Vedic reciters, as a rule, trained from young, first to recite the mantras (Vedic words) correctly, that is, the continuous recitation (saṁhita pāṭha), word for word recitation (pada pāṭha), and rhyming couplets (krama pāṭha). Only after mastering this, they were allowed to go on to learn the 8 complex methods of recitation by way of mode and pitch (such as normal and reverse sequence, like playing arpeggios in music training).479 Such methods prevent errors in recitation and transmission, without the reciter even knowing the meaning of what is being recited. What is preserved is the word, the text.

6.2.2.4 In the Buddhist recital tradition, the specialists begin their training as adults. The nature of the Pali texts—such as following rules of sandhi (euphonic combination), textual patterns [3.3], the use of pericopes [4.2.7], and so on—help in the memorization of the texts [4.2.7]. However, the stress is less on the memorizing, but more on the understanding of the texts recited. Hence, the chances for textual errors are more likely to occur. Such errors are easily corrected in 3 ways. The first is in the group recital itself. On account of the number of reciters, the trend of self-correcting works naturally and easily. Secondly, when the reciters understand what he is reciting; hence, the familiarity with the Dharma terms and passages helps in keeping the recitation correct. The third way is a later development, with the written texts, with which the recitation is checked; or where a prompter follows the recital line by line on the printed text. This last method is now a standard practice during the Pātimokkha recital for monastics.480

6.2.3 Monastics: forest and urban?

6.2.3.1 The question that now arises is: Who were the reciters (bhāṇaka) or those who memorized the Pali texts—the nikāyas to be specific—so that they have been handed down right to our times? More correctly, we should phrase the question more specifically as follows: Who memorized and preserved the ancient texts (the nikāyas, or more broadly, the Ti,piṭaka) until they were written down in Sri Lanka in 100 BCE?481 By the time we have these nikāya texts, the monastic community in India at least comprised traditionally of 2 main divisions: the forest-dwellers (arañña,vāsī) and the community dwellers (gāma,vāsi). The forest-dwellers (arañña,vāsī) or eremites were not merely those who led solitary lives in the forest or some secluded spot, but their main practice was meditation, and led much simpler lives than the community dwellers or coenobites.

476 See L R Goonesekere, “Bhāṇaka,” 1966 (Ency Bsm 2: 688-690). 477 Analayo 2007:17-20. 478 This is however a late term: Mahv 33.102. 479 See eg https://en.wikipedia.org/wiki/Vedic_chant. 480 See Analayo 2007:8 f. 481 A K Warder, Ñāṇamoli’s Path of Discrimination Intro, 1982:xxxix; Gethin, The Buddha’s Path to Awakening, 1992:8; Harvey, Introduction to Buddhism, 1990:3.

98 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

The community dwellers (gāma,vāsī) did not dwell in villages, towns or “among the houses” (antara,- ghare),482 but as prescribed by the Vinaya, dwelled as a community near an inhabited area (village, market town, city) where their noise could not be heard and not too far away for finding alms. The texts define “forest” (as suitable for monastic dwelling):483 “except the village and its precincts, all is forest” (V 3:46);484 “500 bow-lengths distant” (V 4:183);485 “having gone beyond the city-gate’s post, all that is forest” (Vbh 251; Pm 1:176).486

6.2.3.2 The notion that monastics who were community dwellers were “text-oriented” or “book-bias- ed” (gantha,dhura) and forest-dwelling monks were “meditation-oriented” (vipassanā,dhura) is extracan- onical.487 The suttas, for example, recounts how monks who were meditators (jhāyī) conflicted with those who were “Dharma specialists” (dhamma,yoga), implying that they were co-residents of the same community or were in close contact with one another. In fact, nowhere in the suttas is there any statement that the two “tasks”—Dharma training and meditation—are ever separate. Rather they form the integral threefold training in moral virtue, concentration and wisdom [1.1.2.2; 5.4.5.2]. All monastics are supposed to learn and master meditation, that is, some level of mental concentration, including everyday mindfulness. [5.4.5.2]

6.2.3.3 Considering the lifestyle of the kinds of monastics—the forest-dwellers and the community- dwellers—the latter monastics are more likely to be socially engaged, in the sense of accepting group alms-offerings, giving public Dharma talks, conducting communal activities (especially daily pujas) and so on, and to be devoted to sutta recitation and memorization. The task of memorizing the texts is ideally done in a quiet place free from distraction from others: the solitary life of the forest dweller is, in fact, ideal for this. Moreover, with meditation, his mind is better primed for remembering what he recites, and to understand them, too, as part of his practice. It should also be remembered that meditation (by this time, anyway) is also based on the texts, with the practical guidance of meditation teachers who are also well trained in the texts.

6.2.3.4 In Sri Lanka today, where secularization has swept across and flooded most of the temples and centres, so that it is common for the lay Buddhists there to speak of “Vinayaless monks,” those who neither hold the fortnightly Pātimokkha conclave nor bother about the Vinaya rules at all.488 In fact, despite the academic qualifications of many of such modern priests, many, if not most, of them are not experts in the suttas, not to mention the Vinaya.489

482 V 4:176,18’; SA 2:34,28-31. A number of rules in the Pāṭimokkha prescribe the decorum for monastics when “amongst houses,” found mostly in the Sekhiya,dhammas (the last section of Ptmk) (V 4:186,8 f + 19 f + 29 f, 187,4 f + 16 f + 27 f, 188,4 f + 17 f + 28 f + 35 f, 189,9 f + 27 f, 199,15 f): see Pruitt & Norman, The Pātimokkha, 2001 index: antaraghare. 483 All the foll refs are at Vism 2.49-51/72. 484 Araññaṁ nāma ṭhapetvā gāmañ ca gām’upacārañ ca avasesaṁ araññāṁ nāma (V 3:46,30 f). 485 Āraññakaṁ nāma senāsanaṁ pañca,dhanu,satikaṁ pacchimaṁ (V 4:183,6 f). 486 Araññan’ti nikkhamitvā bahi,inda,khilā sabbam etaṁ araññaṁ (Vbh 251,17,14 = Pm 1:176). On inda,khīla, see SD 39.7 (1.2.1.2). 487 All refs to this duality are comy: AA 1:313; DhA 1:8, 68, 154; ThaA 2:101; ApA 275; VA 3:561. 488 This troubling secular trend is also clearly evident in Sinhala monks who work overseas running their own house viharas or being employed as salaried in-house priests in Buddhist Temples and Societies, taking meals during the forbidden hours, even given their own cars to drive around in. 489 The founding Chief High Priest of the missionary Mahavihara in Malaysia, eg, was well known for his writings and talks that employed the populist Dale Carnegie approach, such as “Why worry?” “How to overcome your difficulties?” “Great personalities on Buddhism.” Interestingly, he wrote “Meditation: The only way,” yet, he himself http://dharmafarer.org 99 SD 58.1 The Oral Tradition of the Early Buddhists

Scholars researching on modern Buddhism are struck by “the high literary productivity of the forest- dwelling monks of Sri Lanka in the 20th century, but Tambiah has noted how throughout that country’s history, forest hermitages have produced scholars of great distinction.”490 In other words, it is the dedica- tion and attainment of the Dharma-spirited forest monks who not only preserved the Pali canon for us, 491 but also educate us with its insightful explanations.

6.2.4 The 5 nikayas

6.2.4.0 Before the Common Era, the early Buddhist texts were preserved and propagated by word of mouth. However, as such transmission of steadily collections grew with open canon, it was necessary to introduce certain principles of order and taxonomy. In time, the collection evolved from the dhamma,vinaya of the Buddha’s time to the ti,piṭaka of the modern age. The Tipiṭaka scheme of arranging texts grew into the 3 (ti) baskets (piṭaka) of discipline (Vinaya), discourses (sutta), and scholastic dogmas (Abhidhamma). All Buddhist schools whose literature has been preserved divided their sutta collection into sections called Nikāya or Āgama. Neither term is school-specific; the notion that the Theravāda school used the term nikāya while other schools used āgama is attested neither by Pāli nor Sanskrit sources.492 As for the early Buddhist texts, they are preserved, for the most part, in the 5 nikāyas of the Pali canon, that is, in the Dīgha,nikāya, the Majjhima,nikāya, the Saṁyutta,nikāya, the Aṅguttara,nikāya and the Khuddaka,nikāya.

6.2.4.1 THE DĪGHA,NIKĀYA This Pāli title, which means “the collection of long teachings,” which is the 1st division of the Sutta Piṭaka, the basket of suttas, the “threads” of teachings that go back to the Buddha himself. It is comprised of 34 long suttas, arranged rather arbitrarily into 3 major sections: “the morality section” (sīla-k,khanda), comprising suttas 1-14; “the great division” (mahā,vagga), comprising suttas 14-23; and on Pāṭika (pāṭi- ka,vagga), comprising suttas 24-34. Among the suttas of this Nikāya are such renowned and influential texts as the Aggaññā Sutta (D 27), SD 2.19, the Mahā,parinibbāna Sutta (D 16), SD 9 [6.2.4.2], the Sāmañ- ña,phala Sutta (D 2), SD 8.10, and the Mahā Sati’paṭthāna Sutta (D 22), SD 13.2. Its Pāli Commentary, the Sumaṅgala,viḷāsinī, was composed by Buddhaghosa during the 5th century CE. These “long discourses” have parallels in the Sanskrit Dīrghâgama (DĀ) (all but 3 of its 30 sutras have their parallels in Pāli). Fragments of the Sanskrit recension, which is associated with the Sarvāstivāda or its Mūlasarvāstivāda offshoot, were uncovered in Afghanistan in the 1990s. Before that rediscovery, only a Chinese translation of the Dīrgh’āgama survived, which was attributed to the Dharmaguptaka school, the translation was finished in 413 CE [5.4.5.4(1)]. Although all 3 recensions of this collection have a tripartite structure, only the first section of the Pāli, the sīla-k,khanda, has a counterpart in the Sarvāstivāda and Dharmaguptaka recensions. [5.4.5.4(1)]493 R O Franke noticed that the Dīgha seems to be sort of a propaganda text (“Propagandaschrift”),494 which, Joy Manné adds, serves for “the integration of new monks into the community and into the prac-

neither meditated nor ever taught it. He was extremely popular with the pragmatic lay Buddhists who found suttas and Buddhist teachings otherwise “deep and difficult.” 490 Tambiah, The Buddhist Saints of the Forest, 1984:56 f, passim; qu in P Harrison 2003:131. 491 See eg Ñāṇananda’s Nibbana, the mind stilled, vols 1-3 (2003-2005). 492 J-U Hartmann, Ency of Buddhism 2003: Āgama/Nikāya. On “Canons and literature,” see Routledge Ency of Buddhism (ed Keown & Prebish) 2007:195-205. 493 For details on the Dīgha Nikāya, see Ency Bsm (Sri Lanka) 4: sv; Norman 1983b:32-44. 494 R O Franke, “Das einheitliche Thema des DīghaNikāya,” WZKM 27, 1913b:201.

100 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists tice”: we see here “all of the technicalities of the Teaching appear here in detail” and “sermons on problems connected with the practice and its difficulty.” The Dīgha, then, works like a prospectus to attract the non-Buddhist with what Buddhism has to offer as a “collection of publicity material” for the purpose of conversion.495 Marasinghe, too, observes that “the majority of the discourses of this collection are ... either directly addressed to the ordained disciples ... or are otherwise intended for them.” (2002a:565)

6.2.4.2 THE MAJJHIMA,NIKĀYA This Pali title means “the collection of middle-length teachings,” which is the 2nd of the 5 divisions of the Sutta Piṭaka. It contains 152 suttas divided into 3 major sections (paññāsaka, “the 50s”), with 50 suttas in each of the first two parts and 52 in the third. Each one of these parts is further subdivided into 5 sections (vagga). The suttas are not arranged in any particular order, but suttas with broadly related themes (eg, the 6 sense faculties), similar styles (eg, suttas that contain a shorter, and often verse, summary of doctrine followed by longer expositions) or target audiences (eg, discourses to householders, monks, wanderers, or brahmins) are grouped together in the same section. Some of the earliest canonical accounts of the Buddha’s awakening are found in several of the suttas in this nikāya. The most significant of these is the Ariya,pariyesanā Sutta (26), SD 1.11, which, however, does not include the famous episode of 4 sights. Instead, the Sutta says: “While still young, a black-haired young man endowed with the blessing of youth, in the prime of life, though my mother and father wished otherwise and wept with tearful faces, I shaved off my hair and beard, put on the saffron robe, and went forth from the household life into the homeless life.” (M 26,14/1:163)496 Not all of the suttas in this nikāya are spoken by the Buddha; Ānanda, for example, delivers the Go,- paka Moggallāna Sutta (M 108), SD 33.5, after the Buddha’s passing into parinirvana. The Commentary on Majjhima Nikāya, written by Buddhaghosa is called Papañca,sūdanī. In terms of function, while the Dīgha provides the non-Buddhist with a wide range of remarkable aspects of the Buddha’s teachings, the Majjhima is focused on providing the converts and neophytes “with the fundamentals of the Teaching and the Practice.”497 Hence, we see an overlap between 2 suttas from the Dīgha and the Majjhima: the Sati’paṭṭhāna Sutta (M 10) also appears as the first half of the Mahā Satipaṭṭhāna Sutta (D 22). In this case, the Majjhima version is the shorter original version which works as an instruction text, while D 22 interpolates the Sacca Vibhaṅga Sutta (M 141), SD 11.11, a detailed exposition on the 4 noble truths like an impressive anthology. [6.2.4.1]

The Sanskrit parallel of the Majjhima Nikāya is the Madhyam’āgama (MĀ), which is the Sarvāstivāda school’s recension of this nikāya. This Āgama is only extant in Chinese translation, comprising 222 sutras, of which 98 correspond to Pali suttas found in the Majjhima Nkāya, 8 appear in the Aṅguttara Nikāya, 12 in the Dīgha Nikāya, and 11 in the Saṁyutta Nikāya. [5.4.5.4(2)]498

6.2.4.3 THE SAṀYUTTA,NIKĀYA Saṁyutta means “yoked or grouped together” or “connected with one another”; hence, the Pāli name means “the collection of connected teachings” and is the 3rd of the 5 Nikāyas of the Sutta Piṭaka. It has parallels in the Saṁyukt’āgama of the Sarvāstivāda and Kāśyapīya schools, now extant only in its Chinese translations. The Pāli recension comprises some 2,872 individual suttas. Because of questions as to what

495 Manné, “Categories of suttas in the Pali Nikāyas … ,” JPTS 15, 1990:79-81 496 This famous pericope is also found elsewhere: D 4,6/1:115 = 5,6/1:129 = M 26,14/1:163 = 36,13/1:240 = 85,11/1:93 = S 1.20,4/1:9. For more refs, see SD 51.15 (1.3.1.2). 497 Manne JPTS 15 1990:81. 498 For details on the Majjhima Nikāya, see Ency Bsm (Sri Lanka) 6: sv; Norman 1983b:46-49. http://dharmafarer.org 101 SD 58.1 The Oral Tradition of the Early Buddhists constitutes a sutta in this case (some are only one sentence in length), enumerations of the number of suttas in the various saṁyutta/saṁyukta collections vary widely, from just under 3,000 to over 7,000 (the longer of the two Chinese recensions contains 1,362 sutras).499 The Saṁyutta Nikāya is divided into 5 chapters (vagga), which are subdivided into 56 saṁyuttas or books, arranged broadly by subject matter.500 The collection derives its title from the way its teachings are categorized. The 5 vaggas are devoted to: (1) verses (sa,gātha), suttas that in the majority of cases contain verses; (2) causality (nidāna), suttas that deal primarily with epistemology and psychology; (3) the aggregates (khandha), suttas on the 5 aggregates; (4) the 6 sense-bases (saḷ-āyatana), suttas dealing with the 6 bases of consciousness; and (5) the great division (mahā,vagga), which contains suttas on the noble eightfold path (ariya aṭṭh’aṅ- gika magga), the training is moral virtue, in meditation and wisdom. [5.4.5.4(3)] The Commentary on the Saṁyutta Nikāya, written by Buddhaghosa is called Sār’attha-p,pakāsinī.501 Unlike the Dīgha [6.2.4.1] and the Majjhima [6.2.4.2], the Saṁyutta rarely includes a narrative background to the teachings, which are presented directly, and often as a cross-section of various aspectsof a particular teaching or related sets, especially in the 5th, last and longest chapter, on the 7 sets (constituting the 37 limbs of awakening), which are the essence of the Buddha’s teachings.502

6.2.4.4 THE AṄGUTTARA,NIKĀYA The Pāli name means “the collection of numerical teachings,” which is the 4th division of the Sutta Piṭaka. This collection, which may date from as early as the 1st century BCE, consists of 2,198 suttas (9,557 suttas, if we count all the repetitions in the peyyālas, cycles), arranged in 9 sections (nipāta). The suttas in the Pāli collection are arranged sequentially in numbered lists according to their subject matter, beginning with discussions of “ones,” such as nirvana, and progressing up to sets of 11. Much of the Aṅguttara teachings are found elsewhere in the other Nikāyas, and many teachings are simply stated or abridged,503 which assumes our knowledge of them as taught elsewhere, especially the Majjhima and the Saṁyutta. Furthermore, the numerical arrangement of teachings and generally simpler approach clearly shows they are meant very much for the benefit of the lay audience. The Aṅguttara’s commentary, the Manoratha,pūraṇī, was composed by Buddhaghosa during the 5th century CE.504 The Aṅguttara has some close parallels in the Ekottar’āgama (EĀ), extant only in Chinese translation (of unidentified affiliation), but which is much smaller with only 473 sutras [5.4.5.4(4)]. The Oxford Dict- ionary of Buddhism notes that “there is considerable disparity between the Pāli and the Sarvāstivādin versions [of the Aṅguttara], with more than two-thirds of the sūtras found in one but not the other compilation, which suggests that much of this portion of the Sūtra Piṭaka was not formed until a fairly late date.”505

499 See https://en.wikipedia.org/wiki/Sa%E1%B9%83yutta_Nik%C4%81ya. 500 The Saṁyutta is conveniently referenced by saṁyutta and sutta number, eg. Dhamma,cakka Pavattana S is S 56.11: in the 56th saṁyutta, sutta 11. 501 For details on the Saṁyutta Nikāya, see Ency Bsm (Sri Lanka) 7: sv; Norman 1983b:49-54. 502 See S:B 1485-1492. 503 Eg, the 17 wishes of a monastic listed in Ākaṅkheyya S (M 6), SD 59.1, are reduced to only 10 in Ākaṅkha S (A 10.71), SD 82.12, and has no conclusion like M 6. See SD 59.1 (1.1.5.2). 504 For details on the Aṅguttara Nikāya, see Ency Bsm (Sri Lanka) 1: sv; Norman 1983b:54-57. 505 Oxford DB 2003: Ekottara Āgama.

102 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

6.2.4.5 THE KHUDDAKA,NIKĀYA The Pāli name, meaning “the Collection of Minor or Miscellaneous teachings,” refers to the 5th and last division of the Pali Sutta Piṭaka. Such miscellanies or “mixed basket” (Skt ksudraka,piṭaka), were known to have existed in several of the 18 early Buddhist schools [5.4.5.6], including the Dharmaguptaka, Mahāsaghika, and Mahīśāsaka, but none of these recensions are extant (and there is no specific analogue in the Chinese Āgama translations). The Pāli Khuddaka Nikāya is composed of 15 independent books, most of them in verse, and some of them representing the earliest strata of the Pāli canon, others relatively late compositions. The works are generally in verse, comprising the following:506

(1) Khuddaka,pāṭha (Khp), “the minor readings” of 9 basic texts extracted from the canon; said to have been compiled in Sri Lanka, and interpolated into the modern Tipiṭaka.507 (2) Dhammapada (Dh), “word of the Dharma,” an anthology of 423 verses in 26 chapters.508 (3) Udāna (U), “verses of uplift,” 80 suttas recounting the Buddha’s utterances of uplift (udāna) in verse, and the occasion recounted in prose, arranged in 8 chapters.509 (4) Iti,vuttaka (It) “thus said verses” 112 suttas in 4 sections (nipāta), transmitted by the laywomen Khuj- j'uttarā.510; (5) Sutta Nipāta (Sn), “the collection of teachings,” ancient discourses mostly in verse in 5 chapters (vagga).511 (6) Vimāna Vatthu (Vv), “mansion stories,” 85 stories, in 7 chapters (vagga), of devas living in magnifi- cent divine mansions on account of their good karma, related by Moggallāna, Vaṅgīsa and other arhats.512 (7) Peta Vatthu (Pv), “preta stories,” 51 stories, in 4 chapters, of pretas (departed beings), suffering on account of their misdeeds, as recounted by the elder Moggallāna.513 (8) Thera,gāthā (Tha), “elders’ verses,” verse accounts of the 264 elder monks’ lay life of tribulation before their renunciation and ensuing joy; 1279 verses in 21 chapters;514 (9) Therī,gāthā (Thī), “elder nuns’ verses,” verse accounts of 73 elder nuns’ lay life of tribulation before their renunciation and ensuing joy; 522 verses in 73 chapters;515 (10) Jātaka (J), “past-life stories (of the Buddha)” based on Indian fables; only the verses are canonical; the stories commentarial.516 (11) Niddesa (Nd), comprising Maha,niddesa (Nm) and Cūḷa Niddesa (Nc); Nm is a canonical commentary on the Aṭṭhaka,vagga of Sn; Nc is a canonical commentary on the Pārāyana,vagga and the Khagga,- visāṇa Sutta (Sn 1.3).517 (12) Paṭisambhidā,magga (Pm), “the path of analytic wisdom,” an Abhidhamma work attributed to the elder Sāriputta, deals with the understand of the truth as taught by the Buddha.518

506 For details on the Khuddaka Nikāya, see Ency Bsm (Sri Lanka) 6: sv; Hinüber 1996:41-43. 507 On Khp, see Norman 1983b:87 f; Hinüber 1996:43 f. 508 On Dh, see Norman 1983b:58-60; Hinüber 1996:44 f. 509 On U, see Norman 1983b:60 f; Hinüber 1996:46 f. 510 On It, see Norman 1983b:61 f Hinüber 1996:47. 511 On Sn, see Norman 1983b:63-70 Hinüber 1996:48-50. 512 On Vv, see Norman 1983b:70 f; Hinüber 1996:50. 513 On Pv, see Norman 1983b:71 f; Hinüber 1996:50 f. 514 On Tha, see Norman 1983b:72-75; Hinüber 1996:51-54 515 On Thī, see Norman 1983b:75-77; Hinüber 1996:51-54. 516 On J, see Norman 1983b:77-84; Hinüber 1996:55-58. 517 On Nd, see Norman 1983b:84-87; Hinüber 1996:58 f. 518 On Pm, see Norman 1983b:87-89; Hinüber 1996:59 f.

http://dharmafarer.org 103 SD 58.1 The Oral Tradition of the Early Buddhists

(13) Apadāna (Ap), “the harvest of past lives (of the arhats),” relates in verse the noble deeds of the early saints in their past lives; 55 chapters on the lives of 547 elders, and 4 chapters on the lives of the 40 nuns, all recounting how they met a past buddha.519 (14) Buddha,vaṁsa (B), “the lineage of buddhas,” opens with a description of the jewel walk, followed by an account of Sumedha (when Gotama first aspires to buddhahood); then 24 chapters on the 24 past buddhas, ch 26 on Gotama; ch 27, on 3 additional buddhas before Dīpaṅkara; and ch 28 on the distri- bution of the buddha relics.520 (15) Cariyā,piṭaka (C), “the basket of conduct,” 35 jātaka-type stories of the Buddha’s past lives in 3 chapters, on the 10 perfections.521

The Burmese recension of the Pāli canon adds to the collection four other works, totalling 19 books, thus:

(16) Milinda,pañha (Miln), Milinda’s questions.522 (17) Sutta,saṅgaha (Suts), compendium of suttas.523 (18) Peṭakôpadesa (Peṭk), the Piṭaka disclosure.524 (19) Netti-p,pkaraṇa (Nett), the guide.525

Apparently, the Khuddaka serves as addenda of canonical texts that have not been included in the 4 nikāyas (such as the Sutta Nipāta and Dhammapada), and also an appendix for late works to reach out to worldly trends or answer external challenges (such as the Cariyā,piṭaka).

7 Living Buddha Dharma fully

7.1 The middle way of truth and beauty Today, we have such easy access to the early Buddhist texts, with better new translations—including widely available commercial publications, as never before. Naturally, we would feel drawn to return to our spiritual roots and to discover the historical Buddha and his teachings. We feel a profound need to recon- nect with our ancient heritage that we almost lost through a history of rising religious materialism and the vicissitudes of history. Much of the Buddhism we see today are the sweet Galatea crafted by sharp-eyed Pygmalions, the hollow Frankenstein of status-minded Professionals, the petrifying Chimera of narcissistic Gurus baiting for an adoring fandom of believers and benefactors for their heaven on earth. Religion is the oldest, effica- cious tool in the relentless battle for the minds of the masses. Our singular task today, then—like the sharp minds and keen hearts lighting the dark, pre-scientific days—is to educate ourself in every way in terms of the truth of learning and the beauty of the discovery, so that we are not frozen in the amber of religious dogmas, or pickled in material convenience. We must seek the middle way to return to the Buddha of history and his historic teaching of mental liberation that is vitally relevant today as it was and will be.

519 On Ap, see Norman 1983b:89-92; Hinüber 1996:60 f. 520 On B, see Norman 1983b:92-94; Hinüber 1996:62. 521 On C, see Norman 1983b:94 f; Hinüber 1996:63 f. 522 On Miln, see Norman 1983b:110-112; Hinüber 1996:82-86. 523 On Suts, see Hinüber 1996:76 f. 524 On Peṭk, see Norman 1983b:108 f; Hinüber 1996:80-82. 525 On Nett, see Norman 1983b:109 f; Hinüber 1996:77-80.

104 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

7.2 The complete sutta

The suttas, the Buddha’s teachings, must be learned and lived, and learned and lived in full. A part of a computer programme will not run: we need the whole programme, and run the whole programme. This is the only way when we study the suttas. For this reason, careful Dharma scholars, teachers and students joyfully embrace the teachings in full. A leading British Buddhist scholar and practitioner, L S Cousins, in his review of The Middle Length Discourses of the Buddha, a commercial translation of the Majjhima Nikāya by Ñāṇamoli and Bodhi526 (besides his positive points) makes some insightful observations of how we should approach the suttas in the full spirit of the Dharma. In practical terms, Cousins reminds us of the care and standards we must maintain. In his insightful criticism of a very successful commercial translation of the Majjhima Nikāya by Bhikkhu Bodhi, he gives the following feedback (summarized here):

• the translation does not make clear what Pali text or manuscript/s it is using;527 • being largely the work of Ñāṇamoli (died 1960), it does not take much account of recent scholarship; • in order to fit in one volume, it cuts a lot of repetitions, thus altering the text’s emphasis [see below]; • it tends to very uncritically follow traditional commentarial interpretations; • Bodhi, in an effort to keep a standard terminology, has undone some improvements Nanamoli had tried to make. Cousins says that some such translations are “quite unacceptable” and “promulgate widespread misunderstandings,” and that Ñāṇamoli understood this. Cousins says that “what is needed is much more creativity and variety” in translation to try better to convey early Buddhist concepts.

Cousins’s remark on the importance of keeping (and reading) the repetitive passages (or peyyāla) of the Suttas is very relevant here:

Although there are obvious advantages to having the whole work in one volume, especially for the newcomer, the disadvantage is the systematic elimination of repetition. Many readers may see this as a gain but it does amount to the deliberate removal of a meditative element from many suttas. In the original language, when the suttas are chanted rather than read, there is an effect upon the mind which is very much part of their traditional purpose. Moreover, the emphasis is often changed quite critically: when a phrase that is repeated in the original ten times is given only once in translation, it ceases to be a central part which is always retained in memory. The result is that what the sutta tries to stress as important becomes much less noticeable in translation. (Journal of Buddhist Ethics 4 1997: 261 f)

7.3 Discipleship is the path

7.3.1 Most people tend to see the 3 jewels—the Buddha, his teaching and the noble disciples—as some kind of ornament. They separate and cut up these jewels to adorn themselves or store them away like some some thing they possess. The sheen of the jewels fade away, they collect dust and grime in their crowded lives, diminished or devalued like Procrustes truncating or stretching his “guests” on his infamous bed. We do not see the 3 jewels for what they really are, and do not benefit from their light, their beauty, their noble wealth. We are caught up with titled teachers, popular preachers and charismatic charlatans with their populist, secularist, materialist, postmodernist or academic glitter. We are drawn to the looks, voice, holiness

526 1st published 1995, 2nd ed 2001, 3rd ed 2005, 4th ed 2009. 527 Curiously, Bodhi completely omitted any references to the Siamese Tipiṭaka in all his commercial translations. http://dharmafarer.org 105 SD 58.1 The Oral Tradition of the Early Buddhists and truths (teachings),528 and measure them, collect them, show them off529—instead of learning the Dharma and awakening to the path of freedom.530

7.3.2 True faith (saddhā), on the other hand, is the own inner light of goodness that shines to show the goodness of others, especially the Dharma in a teacher, and to connect with that Dharma, so that together they progress on the steps of discipleship to tame our body and speech by moral virtue, know our mind by calm and insight, free ourself with compassion and wisdom. This path is that of listening to the Dharma as heirs of the 3 jewels. By listening, we grow in discipleship: we become true disciples, true hearers (sāvaka) of the Dharma. This discipleship arises from a full living of the Dharma life—in body, speech and mind—mastering the Buddha Dharma in both the letter and the spirit, tasting its good in its beginning, its middle and its end. [1.2.1.3] This is a Dharma-spirited mastery of the suttas so that we can well learn the Buddha’s oral tradition that has been passed down, and pass it on. For the Buddha Dharma to be authentic, we do not need to prove that all the suttas are true, or that everything in them are right. As we work with the suttas, we will discover some things are well learnt, some are badly learnt; some true, some not. With some diligence and reflection, we will be able to discern these things.531 What remains in your sutta quest—as we search the suttas—is a bright, clear, deep and strong light that is our spiritual life. What we seek in the suttas, then, is what is useful in guiding us towards the path of awakening. The only way we can know this is to search the suttas, live the truths and taste the joy of freedom. This is what the Buddha teaches and only this: suffering and its ending. [1.2.2.2]

7.4 Early Buddhism as common sense

7.4.1 The Pali canon or Tipiṭaka in India became the nucleus for what we today call early Buddhism. His- torically, it is not correct to say that Theravāda, that arose in Sri Lanka, is the oldest school, or even has the oldest scripture. Indeed, the term thera,vāda originally meant the “doctrine or school of the elders,” that is, the arhats who convened the 1st council, or better, the “elders” (thera) comprising the Buddha and the early arhats, whose awakening are identical, as attested in the Sambuddha Sutta (S 22.58).532 However, the earliest teachings of the Buddha still extant are more fully preserved in the Pali canon than in another other Buddhist canon today. Since Buddhism is not a book-based religion, it does not really matter even if we are unable to produce ancient manuscripts with Pali teachings. The discoveries of early manuscripts [5.4.5.7] of the ancient texts of ancient schools in Afghanistan or Central Asia do not debunk the fact that the earliest teachings of the Buddha is preserved in the Pali canon.

7.4.2 Even if the Pali suttas, as we have them today, are not the earliest texts, the earliest teachings are embedded in them and easily retrievable from them, unlike the canons or scriptures of other schools extant or extinct. The test is in the pudding itself. The teachings of the Pali suttas show the least influence, importation or modification from external and extraneous sources. Even when the Buddha is recorded as buddhicizing elements from the Vedas or Upanishads, there is always a consistency and unity in his methods and teachings that, for example, reflect the 7 sets [5.2.8.3]. If we are to remove or reject the 7 sets or revise them in a manner that is foreign, even antagonistic, to

528 See Rūpa S (A 4.65), SD 3.14(6) = SD 19.2a (6.5). 529 See Me, the nature of conceit, SD 19.2a. 530 See The teacher or the teaching? SD 3.14. 531 See Sandaka S (M 76), esp §§24-26/1:520 (SD 35.7). 532 S 22.58/3:65 f (SD 49.10).

106 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists the “Pali drift,” the spirit of the Pali canon, such as claiming that the Buddha or an arhat is still not enlight- ened (like a Bodhisattva), then, we have lost our vision of the Buddha and the path.

7.4.3 The spirit of early Buddhism

Even if we are unable to prove “early Buddhism” as a historical reality, we can still live it as a spiritual life. The notion of early Buddhism, the wisdom of one fully awakened, the Buddha, invites us to examine what makes an ideal religion, a truly viable spiritual path that humanizes us so that we are able to attain divinity here and now, and envision the highest good and head that way. The spirit of early Buddhism can be defined as the 3 “essential qualities” of this common-sense “ideal spirituality”:

(1) Impermanence Everything changes: we may deny God or goodness but not impermanence. This is the beauty that shines on all realities. Understanding this frees us from ignorance and suffering. There is no doubt here: only faith or wisdom. This is true reality. The Buddha arises in history as the embodiment of impermanence like the sun that lights up the skies; he passes away, like the setting sun, as the final proof of the true reality that is his teaching. See the Pavāraṇā Sutta (S 8.7), SD 49,11, and the (Saḷ-āyatana) Anusaya Sutta 1 (S 35.58), SD 70.17.

(2) Suffering Suffering is when we look outside for solutions for issues that are really within. As we think so we are. A ritual is a vain pursuit for succour or saviour in another. Salvation and awakening are within. The Buddha awakens and through his teaching, we awaken as arhats; hence, awakening is the same, just as freedom is immeasurable, like the open space we breathe and move in. Suffering follows not so long as we move up the path to the end of time. See the Sambuddha Sutta (S 22.58), SD 49.10.

(3) Non-self We do, we are, we have; we can never identify with any of it. Early Buddhism is the path that does not identify with itself. Tribal Buddhism is self-identity view embodied. Our journey is the breath: take it in, give it back; it is not ours. The breath then becomes the mind of wisdom, the heart of freedom. Every step we take, we leave behind our views of self: this is true renunciation. For, the path is not out there but an inner journey in the Dharma, only we can take ourself, just as the Buddha himself has done, renouncing worldliness, of freeing the world, and arriving in nirvana. See the Gārava Sutta (S 6.2), SD 12.3, and the (Anicca) Cakkhu Sutta (SD 25.1), SD 16.7)

These are the 3 characteristics of true reality; they are also the 3 fetters, breaking which we reach the path or awakening. These are the practical essence of early Buddhism.

— — —

http://dharmafarer.org 107 SD 58.1 The Oral Tradition of the Early Buddhists

Bibliography [For more titles, see SD Guide] Allon, Mark 1997 “The oral composition and transmission of early Buddhist texts,” in Peter Connolly & Sue Hamilton (eds), Indian Insights: Buddhism, Brahminism and Bhakti: Papers from the An- nual Spalding Symposium on Indian Religions. London: Luzac Oriental 1997:39-61. Analayo Bhikkhu 2007 “Oral dimensions of Pāli discourses: Pericopes, other mnemonic techniques and the oral performance context,” Canadian Journal of Buddhist Studies 3 2007: 5-33. download Chinese tr (extract): 巴利口誦傳統的形式與功能 in 第一屆巴利學與佛教學術研討會, Tai- wan: 南華大學世界禪學研究中心, 2006:1-9. http://www.buddhismuskunde.uni- hamburg.de/fileadmin/pdf/analayo/OralDimensions.pdf. 2008 “The verses on an Auspicious Night, explained by Mahakaccana--A study and translation of the Chinese version,” Canadian Journal of Buddhist Studies 4 2008: 5-29. http://www.buddhismuskunde.uni-hamburg.de/fileadmin/pdf/analayo/Verses.pdf. 2009 “The vicissitudes of memory and early Buddhist oral transmission,” Canadian Journal of Buddhist Studies 5 2009:5-19. 2015a “Āgama/Nikāya,” in Brills’ Encyclopedia of Buddhism, (ed) O von Hinüber, Leiden: Brill 1 2015:50-59. 2016 “Āgama and aṅga in the early Buddhist oral tradition.” Singaporean Journal of Buddhist Studies 3 2016:9-37. 2020 “Early Buddhist oral transmission and the problem of accurate source monitoring.” Mind- fulness 11, 2020:2715-2724. https://doi.org/10.1007/s12671-020-01499-z. Brough, John 1950 “Thus have I heard … “ Bulletin of the School of Oriental and African Studies 13,2, 1950: 416-426. DOI: https://doi.org/10.1017/S0041977X00083555 Cousins, Lance Selwyn 1983 “Pali oral literature,” in (eds) O Denwood & A Piatigorsky, Buddhist Studies: Ancient and Modern. London: Centre for South Asian Studies, School of Oriental and African Studies, Univ of London & Curzon Press/Barnes & Noble, 1983:1-11. 1997 Review of The Middle Length Discourses of the Buddha [Majjhima Nikāya tr Ñāṇamoli & Bodhi, 1st ed, Boston: Wisdom Books, 1995] Journal of the Buddhist Ethics 4 1997: 260- 280. http://www.buddhistethics.org/4/cous1.pdf. 2009 “Scholar monks and meditator monks revisited.” In John Powers & Charles S Prebish (eds), Destroying Māra Forever: Buddhist ethics essays in honor of Damien Keown, Ithaca, NY: Snow Lion, 2009:31-46. 2013 “The early developments of Buddhist literature and language in India.” Journal of the Oxford Centre for Buddhist Studies 5, 2013:89-135. Douglas, Mary 2007 Thinking in Circles: An Essay on Ring Composition. Terry Lecture series. New Haven, CT: Yale University Press, 2007. Gethin, Rupert 2004c “On the practice of Buddhist meditation according to the Pali Nikāyas and exegetical sources,” in Buddhismus in Geschichte und Gegenwart 10 (Hamburg), 2004:17-37. Govender, Selva

108 http://dharmafarer.org Piya Tan The Oral Tradition of the Early Buddhists

1992 “Evam me sutam. A critical evaluation and interpretation of oral features of the Brahma- jala Sutta.” MA thesis, Orality-Literacy Studies, Faculty of Humanities, Univ of Natal. Dur- ban, 1992. https://researchspace.ukzn.ac.za/handle/10413/7084. Griffiths, Paul 1983 “Buddhist Jhāna: A form-critical study.” Religion 13,1 Jan 1983:55-68. Harrison, Paul M 2003 “Mediums and messages: Reflections on the production of Mahāyāna sūtras.” Eastern Buddhism 35,1-12 2003:115-151 esp 123. Hartmann, Jens-Uwe 2004a “Āgama/Nikāya,” in Encyclopedia of Buddhism, R E Buswell (ed), New York: Macmillan 2004 1:10-12. McMahan, David L 1998 “Orality, writing, and authority in South Asian Buddhism: Visionary literature and the struggle for legitimacy in the Mahāyāna.” History of Religion 3 1998:249-274. A detailed discussion of the shift from orality to written culture in Buddhist textual transmission. Michelman, Paul 2004 “What leaders allow themselves to know: Warren Bennis unveils some new thinking on the filters that govern decision making,” Harvard Management Update, Feb 2004. Nanayakkara, S K 1990 “Evam me sutam.” Encyclopaedia of Buddhism. Colombo: Govt of Sri Lanka, 1990 5:174 f. Skilling, Peter 2009b “Redaction, recitation, and writing: Transmission of the Buddha’s teaching in India in the early period.” In (eds) Stephen C[arl] Berkwitz, Juliane Schober & Claudia Brown, Buddh- ist Manuscript Culture: Knowledge, Ritual, and Art. London/New York: Routledge, 2009: 53-75. Sujato, Bhikkhu & Bhikkhu Brahmali 2015 The Authenticity of the Early Buddhist Texts. [2013. Chroniker Press.] Perth, Australia, 2015. Repr as Supplement to JOCBS vol 5. Download here or https://ocbs.org/the- authenticity-of-the-early-buddhist-texts-2/ or https://ocbs.org/wp- content/uploads/2015/09/authenticity.pdf. Wynne, Alex 2004 “The oral transmission of the early Buddhist literature.” In Journal of the International Assoc of Buddhist Studies 27,1 2004:97-128. 2005 “The historical authenticity of early Buddhist literature: A critical evaluation.” Wiener Zeitschrift für die Kunde Südasiens/Vienna Journal of South Asian Studies, Band 49 2005: 35-70.

0811101 091219 100917 101222 130513 180222 210119 210721

http://dharmafarer.org 109